Back

First < Newer 41-60 of 137 Older> Last

Latest Forum Topics / OCBC Bank Last:32.39

+0.47

Post Reply

the AI fever globally

chartiskao
Supreme

14-Aug-2026 06:05

chartiskao About Contact Quote this Post!

x 0
x 0

Alert Admin

" while -if" loop for Singapore

China is actually much further along in building an AI governance framework than the US, but its approach is very different from both the EU and the current Trump administration.
The key insight is:

China is not choosing between &ldquo AI development&rdquo and &ldquo AI regulation.&rdquo It is trying to do both simultaneously &mdash accelerate AI nationally while putting the most powerful applications under increasingly tight state control.

As of August 2026, I would describe China' s framework as &ldquo accelerate broadly, control selectively, and tightly manage strategic AI.&rdquo

1. China started regulating AI before ChatGPT

China already had an AI-specific regulatory architecture before the current US debate.
The basic layers are:
Algorithm Recommendation Rules &mdash 2022
&darr
Deep Synthesis Rules &mdash 2023
&darr
Generative AI Measures &mdash 2023
&darr
AI-generated content labelling &mdash 2025
&darr
AI Safety Governance Framework 2.0 &mdash 2025
&darr
AI application security classification &mdash 2026
China therefore hasn' t been waiting for one giant &ldquo AI Act.&rdquo It has been building the system incrementally.

2. The 2023 Generative AI rules were a major step

China' s Interim Administrative Measures for Generative AI Services came into force in August 2023.
The framework puts obligations on providers of generative AI services, including requirements around:

data and training practices
content management
personal information
security
provider responsibility
handling illegal or harmful content
registration/security procedures for certain algorithms and services

This is fundamentally different from the Trump approach described in your previous article.
China' s philosophy is essentially:

You can build AI, but once you put it into society, the government wants to know what it is doing.

3. Then China introduced AI-generated content labelling

This is an underrated development.
China introduced measures requiring AI-generated/synthetic content to be labelled, including both visible and technical identification mechanisms. The measures were promulgated in March 2025.
Think about the implications.
A Chinese AI-generated:

video
image
audio
text

can potentially be traced or identified as synthetic.
That is partly about misinformation and fraud, but it also fits China' s broader philosophy of traceability and accountability.

4. September 2025 was the really important development: AI Safety Governance Framework 2.0

This is where China starts moving toward something resembling a frontier-AI risk framework.
China released AI Safety Governance Framework 2.0 in September 2025.
It moved toward:
risk classification &rarr risk grading &rarr testing &rarr prevention &rarr response
rather than treating all AI applications as equally dangerous.
This is extremely important when comparing China with the Trump debate.
The Chinese approach is increasingly:

The more capable/dangerous the AI, the greater the governance requirements.

That is much closer to what the Pentagon faction in your article is asking for.

5. China is now moving toward AI risk tiers

This is probably the most important direction to watch in 2026.
China' s framework increasingly distinguishes between different levels of AI risk rather than saying:

&ldquo AI = regulated.&rdquo

The government is working toward categorised and tiered management, including risk testing and evaluation systems. China' s 2025 global AI governance proposal explicitly called for risk assessment, targeted prevention, categorised/tiered management and AI safety testing.
And in July 2026, China released a draft national standard for AI application security classification and grading.
That is significant.
It means China is moving from:
&ldquo We have AI rules.&rdquo
toward:
&ldquo We classify AI according to its risk and apply different controls.&rdquo

6. But here is the contradiction: China is simultaneously pushing AI MUCH harder

This is the part many Western discussions miss.
China' s AI policy isn' t simply:

Regulation = slow AI.

Quite the opposite.
In August 2025, the State Council launched the AI Plus Action Plan.
It calls for extensive AI integration into:

science
manufacturing
consumer services
public welfare
governance
security
international cooperation

and aims to expand AI throughout the Chinese economy.
So Beijing' s formula is:

Regulate AI + massively deploy AI

rather than:

Regulate AI OR deploy AI

That' s an important distinction.

7. China is also pushing AI into physical industries

This is where China' s strategy could become extremely powerful.
China isn' t just trying to create ChatGPT competitors.
It wants AI embedded into:
Factories
&rarr Robotics
&rarr Cars
&rarr Humanoids
&rarr Consumer electronics
&rarr Healthcare
&rarr Finance
&rarr Government
&rarr Military/security
&rarr Industrial production
For example, in June 2026 China' s Ministry of Commerce announced 17 measures to integrate AI into consumption, including intelligent consumer electronics and humanoid robots.
That tells you something about Beijing' s strategy.
China sees AI as a productivity revolution, not merely a software industry.

8. And now China is becoming more national-security focused

This is where the comparison with Trump becomes fascinating.
Reuters reported in July 2026 that Beijing was considering restrictions on overseas access to China' s most advanced AI models.
The discussions reportedly involve companies including:

Alibaba
ByteDance
Z.ai

and potentially restrictions on foreign access to future advanced models.
That' s a major philosophical change.
China is effectively asking:

&ldquo Should our most powerful AI models be treated like ordinary commercial software?&rdquo

Increasingly, the answer appears to be no.

9. China is also restricting strategic AI assets from leaving the country

This is even more important.
In June 2026, China introduced comprehensive rules concerning outbound investment involving sensitive technology sectors, including AI.
The rules allow greater government scrutiny over:

AI technology transfers
overseas deals
strategic assets
technical personnel
sensitive technology

and came amid national-security concerns surrounding Chinese AI assets moving overseas.
So Beijing is moving toward something resembling:

AI = strategic national asset

rather than simply:

AI = private-sector technology

10. And China is now regulating AI companions

This sounds trivial compared with frontier AI.
It isn' t.
In July 2026, China' s new rules governing AI companion/emotional-interaction services came into force.
They address risks such as:

emotional dependency
psychological manipulation
unhealthy reliance
human-like emotional relationships

Major Chinese platforms subsequently shut down or modified some AI companion services.
This demonstrates something distinctive about China' s system:
Beijing is willing to regulate AI applications based on perceived social consequences, not merely technical capability.

So compare China with Trump

This is where your previous article becomes much more interesting.

	🇨 🇳 China	🇺 🇸 Trump administration
Basic philosophy	Control + accelerate	Accelerate + selectively control
AI regulation	Extensive existing framework	Still developing
Frontier AI	Increasingly risk-tiered	Debate between accelerationists & Pentagon
Content	Strong government controls	Concern about censorship
AI safety	Government-led	Government/private-sector debate
AI deployment	Aggressively encouraged	Aggressively encouraged
State authority	Very strong	Federal vs state conflict
Strategic AI	Increasingly treated as national asset	Increasingly viewed through national security
Overseas AI access	Increasing restrictions	Increasing export controls
Industrial AI	Huge priority	Huge priority
Open source	Encouraged strategically	Strong private-sector ecosystem
AI companions/social risks	Direct intervention	Much less developed

The really interesting convergence

Here' s the part I think you should pay attention to.
Trump' s America and Xi' s China are actually moving toward each other on frontier AI.
Not ideologically.
But strategically.

America:

&ldquo We must win the AI race.&rdquo
&darr
&ldquo But some AI capabilities may be national-security risks.&rdquo
&darr
&ldquo Maybe the most powerful models need testing/control.&rdquo

China:

&ldquo We must accelerate AI deployment.&rdquo
&darr
&ldquo But AI is a national-security asset.&rdquo
&darr
&ldquo The most powerful models may need special controls.&rdquo
That is a surprisingly similar destination.

But their systems are fundamentally different

China has one enormous advantage in this area:

It doesn' t have the US federal-versus-state problem.

Beijing can decide:

&ldquo This category of AI requires testing.&rdquo

and implement it nationally.
Trump' s administration faces:
Federal government
vs
California
vs
Texas
vs
New York
vs
Congress
vs
AI companies
vs
Pentagon
That makes America' s regulatory process much slower and more politically contentious.
Reuters reported this week that the Trump administration is continuing its push for federal AI pre-emption while states are developing their own AI rules.

My bigger conclusion

I wouldn' t describe China as having &ldquo more AI regulation&rdquo and stop there.
I' d describe its strategy as:

Build the world' s largest AI ecosystem while constructing a Chinese state-controlled risk-management layer around it.

And that produces a very different model from Europe.

🇪 🇺 Europe

Safety &rarr regulation &rarr deployment

🇺 🇸 Trump

Innovation &rarr deployment &rarr selective security controls

🇨 🇳 China

Innovation + deployment + state control simultaneously
And that' s why China could become the most interesting competitor to the US in AI over the next 5&ndash 10 years.
The question isn' t simply whether DeepSeek, Qwen, Alibaba or ByteDance can produce a better model.
The much bigger question is:

Can China combine frontier models + cheap compute + robotics + manufacturing + enormous industrial deployment + government coordination into one integrated AI economy?

If it can, then the AI race isn' t simply OpenAI vs DeepSeek.
It becomes:
🇺 🇸 Silicon Valley + Nvidia + hyperscalers
vs
🇨 🇳 Alibaba + Tencent + ByteDance + Huawei + DeepSeek + China' s semiconductor/robotics/manufacturing ecosystem.
And China' s AI Plus strategy suggests Beijing is explicitly trying to turn AI into an economy-wide productivity system, rather than just winning the chatbot race.

chartiskao ( Date: 14-Aug-2026 06:02) Posted:

how sg will play the global AI' s " Trump" card
The deeper issue is that AI has moved from being primarily an economic/technology race into a national-security problem.

My reading: Trump has not really chosen an AI camp yet

The article describes three factions:

Camp	Main people	Priority	Likely policy
Accelerationists	David Sacks, JD Vance, Silicon Valley allies	Win the AI race against China	Minimal regulation, maximum deployment
National-security hawks	Pentagon / Pete Hegseth	Prevent AI becoming a cyber/warfare weapon	Testing, restrictions, government oversight
Pragmatists	Susie Wiles, Scott Bessent	Preserve AI investment while controlling catastrophic risks	Voluntary standards + targeted regulation

The important point is that Trump himself appears to be sitting between these camps.
That makes sense politically.
Trump wants:
AI dominance &rarr investment &rarr stock-market gains &rarr American technological supremacy &rarr China containment
But if frontier AI can genuinely discover zero-day vulnerabilities or conduct sophisticated cyber operations, the equation changes:
AI dominance = national-security vulnerability as well as economic advantage.

The Anthropic Mythos episode is the turning point

This is the most important part of the article.
Before Mythos, the administration could reasonably say:

&ldquo AI safety is mostly an excuse to regulate American companies.&rdquo

But if an AI system can autonomously identify software vulnerabilities at extraordinary speed, the government has a completely different problem.
Imagine a frontier model becoming capable of:

finding vulnerabilities in critical infrastructure
identifying weaknesses in military systems
discovering zero-days
generating sophisticated malware
automating cyber reconnaissance
assisting biological or chemical research
operating continuously without human fatigue

Then the traditional Silicon Valley argument of &ldquo move fast and break things&rdquo becomes much less attractive to the Pentagon.
This is why Anthropic' s decision not to release the model may have been more politically important than another year of AI-safety lobbying.
The technology itself changed the argument.

This creates a fascinating Trump contradiction

Trump' s original AI strategy was essentially:
Deregulate America &rarr accelerate innovation &rarr beat China.
But the national-security argument is:
Accelerate America &rarr create extremely powerful systems &rarr potentially create weapons that adversaries can exploit &rarr therefore impose controls.
So the same objective &mdash beating China &mdash produces two contradictory policies.

Accelerationist logic

We need to move faster because China is catching up.

Pentagon logic

We need to control the most dangerous capabilities because China is our adversary.

Both sides are actually arguing from national security.
That' s why this debate is much harder to resolve than ordinary regulation-versus-deregulation politics.

And this is where &ldquo analysis paralysis&rdquo becomes dangerous

The administration has effectively created a policy contradiction:

Policy A

Federal government should prevent states from imposing AI regulations.

Policy B

Federal government increasingly believes certain AI capabilities may require restrictions.
Those two ideas don' t necessarily coexist comfortably.
Suppose California says:

&ldquo Frontier models above capability X must undergo security testing.&rdquo

Trump' s federal government says:

&ldquo No, AI regulation must be nationally uniform.&rdquo

But then Washington itself says:

&ldquo Actually, frontier models capable of autonomous cyber exploitation need government evaluation.&rdquo

Now the obvious question becomes:
Why should Washington be allowed to regulate AI security while California isn' t?
That is the political/legal tension the article is identifying.

The really important test is NOT the announcement

The article gets this exactly right.
Don' t focus too much on Trump' s speeches.
Don' t focus too much on the names of the committees.
Don' t even focus too much on whether Washington says &ldquo AI safety.&rdquo
Look at three concrete things.

1. Is evaluation mandatory?

This is the biggest dividing line.
Voluntary submission
&rarr accelerationists retain the upper hand.
Mandatory evaluation before deployment
&rarr national-security hawks have won a major concession.
Government licensing
&rarr this would represent a much more dramatic break from Trump' s original philosophy.

2. What happens when a company refuses?

This is even more important.
Suppose OpenAI, Anthropic, Google or Meta says:

&ldquo We aren' t submitting this model.&rdquo

What happens?
If the government can say:

&ldquo Then you cannot deploy it.&rdquo

You have genuine regulation.
If the government can only say:

&ldquo Please reconsider.&rdquo

Then it is largely political theatre.

3. Does federal pre-emption survive?

This may ultimately tell us which Trump faction has more power.
If Washington simultaneously says:
&ldquo States cannot regulate AI&rdquo
and
&ldquo The federal government needs stronger AI-security controls&rdquo
then Trump is essentially attempting to centralise AI regulation in Washington.
That could actually produce more federal power over AI, not less.

There is another fascinating angle: China

This is where I think the article could go even further.
The accelerationists' strongest argument isn' t really:

&ldquo Regulation is bad.&rdquo

It' s:

&ldquo The United States cannot afford to slow down while China is racing ahead.&rdquo

That argument is extremely powerful inside Trump' s worldview.
But the Pentagon can turn it around:

&ldquo Precisely because China is racing ahead, we need to know what these systems can do before they are deployed.&rdquo

So the ultimate Trump policy may become:

Low regulation for ordinary AI

but

Heavy controls for frontier AI with military significance.

That is probably the most politically sustainable compromise.

What this means for the AI companies

This is where I would pay particularly close attention.

OpenAI / Google / Meta

They have enormous scale and resources to comply with testing requirements.
So regulation could actually strengthen the incumbents.
That' s the accelerationists' best argument.
If government requires expensive security evaluations, smaller AI companies may simply be unable to compete.
Therefore:
Regulation can paradoxically reduce competition.
That is an extremely important investment consideration.

Anthropic is particularly interesting

Anthropic' s behaviour creates a strange political position.
It effectively demonstrated:

&ldquo We have developed something sufficiently dangerous that we don' t want to release it.&rdquo

That gives the safety camp credibility.
But it also demonstrates something the accelerationists fear:
The company itself can decide what is safe.
That creates the government question:

Should Anthropic decide whether Mythos is safe, or should the Pentagon decide?

That is ultimately what this fight is about.

My probability tree for Trump AI policy

If I were analysing this as an investor rather than a political journalist, I' d assign something roughly like this:

Outcome	My probability	Market implication
Light-touch federal framework	45%	Bullish for AI
Mandatory frontier-model testing	35%	Moderate impact favours big players
Strict licensing regime	10%	Negative for smaller AI companies
Major national-security restrictions	10%	Potential AI-sector shock

I would not expect Trump suddenly to embrace Biden-style broad AI regulation.
That would contradict the economic and geopolitical logic of his administration.
Much more likely is:

Deregulation for the broad AI economy + national-security controls around the most powerful frontier systems.

The bigger investment conclusion

This is where I think the article becomes much more important than it initially appears.
The market has largely treated AI as:
chips &rarr data centres &rarr electricity &rarr cloud &rarr models &rarr applications
But Washington is increasingly treating frontier AI as:
strategic infrastructure + military capability + cyber weapon + economic engine.
That changes the valuation framework.
The next stage of the AI cycle may therefore be less about:

&ldquo Which company has the best chatbot?&rdquo

and more about:

&ldquo Which companies can develop frontier AI while satisfying governments that their systems are safe enough to deploy?&rdquo

That potentially favours the largest, best-capitalised companies.

And this is why I would NOT interpret Trump' s indecision simply as weakness

There is actually a rational reason for the hesitation.
The technology is changing faster than the regulatory framework can adapt.
Trump began with:

Speed first.

Anthropic' s Mythos effectively forced Washington to confront:

&ldquo What if speed itself becomes the security problem?&rdquo

So Trump' s AI policy may be entering its second phase:
Trump 2.0 AI Phase 1:
🇺 🇸 Beat China through deregulation.
Trump 2.0 AI Phase 2:
🇺 🇸 Beat China, but don' t accidentally create an uncontrollable national-security threat.
And the decisive question now is whether Phase 2 remains voluntary or becomes enforceable.
That is the signal I would watch &mdash much more than Trump' s rhetoric.
One final point: because the article is describing events in 2026, some of the specific claims about Mythos, the June order and the August meetings are worth independently verifying before treating them as established facts. If you want, I can deep-dive the actual August 2026 Trump AI power struggle &mdash Sacks vs Hegseth vs Wiles/Bessent &mdash and map out who currently has the most influence, what each person wants, and which AI companies/semiconductor stocks benefit from each outcome.

chartiskao ( Date: 12-Aug-2026 06:45) Posted:

The accelerated global AI capital expenditure cycle has transformed Singapore-listed technology and precision engineering (PE) stocks into primary beneficiaries of the physical hardware build-out.
As global cloud hyperscalers increase infrastructure investments, institutional capital has flowed back into SGX small-and-mid-cap (SMID) tech counters. Net institutional inflows into local tech stocks exceeded S$580 million in the first half of 2026, triggering significant valuation re-ratings across semiconductor equipment manufacturers, test handling specialists, and precision component fabricators.

Core SGX Tech & Precision Engineering Beneficiaries

                             SGX Semiconductor Value-Chain
                                           │ 
         ┌ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┼ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┐ 
         ▼                                  ▼                                  ▼ 
┌ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┐        ┌ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┐        ┌ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┐ 
│      UMS Integration     │        │        AEM Holdings      │        │       Frencken Group     │ 
│    (SGX: 558 / Pure-Play │        │    (SGX: AWX / Advanced  │        │   (SGX: E28 / Diversified│ 
│    Deposition &  Etch)    │        │     Testing &  Packaging) │        │    High-NA EUV Mecha)    │ 
└ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┬ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┘        └ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┬ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┘        └ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┬ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┘ 
             │                                  │                                  │ 
             ▼                                  ▼                                  ▼ 
 Applied Materials Tier-1           Fabless AI / HPC Clients         ASML &  Healthcare / Industrial

1. Pure-Play Front-End Equipment: UMS Integration (SGX: 558)

Value Chain Exposure: UMS produces high-precision components and sub-assemblies for deposition, etch, and chemical mechanical planarization (CMP) tools used in semiconductor fabrication. Its revenue is tied to Applied Materials (AMAT).
AI Capex Catalyst: Unprecedented demand for advanced nodes and high-bandwidth memory (HBM) wafer fabrication equipment has driven strong order book visibility.
Financial & Valuation Profile: Strong balance sheet with a trailing net margin of ~17.4%. Offers attractive dividend yields (~4.4%), serving as a high-income, high-exposure proxy for global foundry equipment capex.

2. Advanced Packaging & AI Accelerators: AEM Holdings (SGX: AWX)

Value Chain Exposure: AEM provides system-level testing (SLT) solutions and automated handling equipment for advanced logic chips, graphics processing units (GPUs), and AI accelerators.
AI Capex Catalyst: As chiplet architectures and 2.5D/3D advanced packaging become standard for AI accelerators, testing complexity rises exponentially. AEM has benefited from a high-volume production ramp for its major fabless AI and high-performance computing (HPC) clients.
Financial & Valuation Profile: High operational beta stock. Q1 2026 revenue expanded 35.8% YoY to S$116.9M. Valuations reflect significant growth optionality as AEM expands testing partnerships beyond traditional lead customers into broader AI foundry environments.

3. Diversified Precision Mechatronics: Frencken Group (SGX: E28)

Value Chain Exposure: Outsourced design and high-precision manufacturing for extreme ultraviolet (EUV) lithography systems (supporting ASML), alongside medical, analytical, and industrial automation equipment.
AI Capex Catalyst: Demand for High-NA (Numerical Aperture) EUV lithography tooling is accelerating as chipmakers scale sub-2nm nodes and HBM architecture.
Financial & Valuation Profile: Offers downside protection relative to pure-play tech stocks due to non-semiconductor revenue streams (medical, industrial) that cushion macro volatility.

4. Specialized Niche & Component Plays

Grand Venture Technology (GVT) (SGX: JLB): Serves front-end semiconductor equipment, aerospace, and medical sectors. GVT is capturing increased market share as Tier-1 equipment makers expand production capacity in Southeast Asia.
Micro-Mechanics (Holdings) (SGX: 508): Manufactures consumable tools and high-precision parts used in semiconductor die-attach and wire-bonding processes. Benefits from high global fab utilization rates.

Key Financial & Operational Metrics Comparison

Stock Name & Ticker	Core AI/Tech Exposure Area	Revenue Growth Drivers	Trailing Net Margin	Dividend Profile	Risk Factor
UMS Integration (SGX: 558)	Deposition & Etch Tool Components	Applied Materials order ramp new Penang factory expansion.	~17.4%	4.0% &ndash 4.5% (Quarterly payouts)	Customer concentration in Applied Materials.
AEM Holdings (SGX: AWX)	System-Level Testing (SLT) for AI Chips	High-Performance Computing (HPC) & AI accelerator testing.	~6.5%	< 1.0% (Prioritizing reinvestment)	Customer concentration high earnings volatility.
Frencken Group (SGX: E28)	EUV Lithography Components & Modules	High-NA EUV rollout 2H26 memory capex rebound.	~4.5%	~1.0% (Stable dividend record)	Margins impacted by non-tech sector drags.
Grand Venture Tech (SGX: JLB)	Ultra-Precision CNC & Sub-assemblies	Front-end fab equipment module outsourcing.	~5.5% &ndash 7.0%	~1.5%	Integration costs from acquired facilities.

Investment Themes & Catalyst Roadmap

1. Operational ASP Increases Lift Profit Margins

Surveys indicate that 41% of precision engineering firms expect higher Average Selling Prices (ASPs) in Q3. This pricing power allows SGX-listed contract manufacturers to pass on rising raw material and energy costs, driving expansion in gross profit margins.

2. Shift from Multiple Expansion to Earnings Realization

While H1 2026 stock price performance was largely driven by valuation re-ratings (higher P/E multiples), H2 earnings expectations rely on deliveries and bottom-line profit execution. Analysts project earnings growth for the technology/industrial sector to remain a primary driver for the broader Straits Times Index (STI).

3. Institutional Inflows & Catalysts

The Singapore Exchange (SGX) and the Monetary Authority of Singapore (MAS) have actively supported market liquidity for mid-cap tech stocks through capital market initiatives and the rollout of the SGX-Nasdaq dual-listing bridge. Increased institutional participation in iEdge Singapore Next50 constituents has improved liquidity across names like UMS, AEM, and Frencken.

Downside Risks to Monitor

Hyperscaler Capex Digestion: If global cloud tech giants slow infrastructure growth in 2027 due to enterprise AI monetization challenges, upstream order cancellations could affect SGX precision component suppliers.
Customer Concentration: UMS (AMAT) and AEM (Intel/key AI fabless) carry high customer concentration risks. Order volatility from primary customers impacts quarterly earnings.
Geopolitical Supply Chain Shocks: Continued friction in the Middle East and US tariff actions pose supply chain risks for chemical feedstocks and international freight costs.

Useful To Me Not Useful To Me

chartiskao
Supreme

14-Aug-2026 06:02

chartiskao About Contact Quote this Post!

x 0
x 0

Alert Admin

how sg will play the global AI' s " Trump" card
The deeper issue is that AI has moved from being primarily an economic/technology race into a national-security problem.

My reading: Trump has not really chosen an AI camp yet

The article describes three factions:

Camp	Main people	Priority	Likely policy
Accelerationists	David Sacks, JD Vance, Silicon Valley allies	Win the AI race against China	Minimal regulation, maximum deployment
National-security hawks	Pentagon / Pete Hegseth	Prevent AI becoming a cyber/warfare weapon	Testing, restrictions, government oversight
Pragmatists	Susie Wiles, Scott Bessent	Preserve AI investment while controlling catastrophic risks	Voluntary standards + targeted regulation

The Anthropic Mythos episode is the turning point

This is the most important part of the article.
Before Mythos, the administration could reasonably say:

&ldquo AI safety is mostly an excuse to regulate American companies.&rdquo

But if an AI system can autonomously identify software vulnerabilities at extraordinary speed, the government has a completely different problem.
Imagine a frontier model becoming capable of:

finding vulnerabilities in critical infrastructure
identifying weaknesses in military systems
discovering zero-days
generating sophisticated malware
automating cyber reconnaissance
assisting biological or chemical research
operating continuously without human fatigue

This creates a fascinating Trump contradiction

Accelerationist logic

We need to move faster because China is catching up.

Pentagon logic

We need to control the most dangerous capabilities because China is our adversary.

Both sides are actually arguing from national security.
That' s why this debate is much harder to resolve than ordinary regulation-versus-deregulation politics.

And this is where &ldquo analysis paralysis&rdquo becomes dangerous

The administration has effectively created a policy contradiction:

Policy A

Federal government should prevent states from imposing AI regulations.

Policy B

Federal government increasingly believes certain AI capabilities may require restrictions.
Those two ideas don' t necessarily coexist comfortably.
Suppose California says:

&ldquo Frontier models above capability X must undergo security testing.&rdquo

Trump' s federal government says:

&ldquo No, AI regulation must be nationally uniform.&rdquo

But then Washington itself says:

&ldquo Actually, frontier models capable of autonomous cyber exploitation need government evaluation.&rdquo

Now the obvious question becomes:
Why should Washington be allowed to regulate AI security while California isn' t?
That is the political/legal tension the article is identifying.

The really important test is NOT the announcement

1. Is evaluation mandatory?

2. What happens when a company refuses?

This is even more important.
Suppose OpenAI, Anthropic, Google or Meta says:

&ldquo We aren' t submitting this model.&rdquo

What happens?
If the government can say:

&ldquo Then you cannot deploy it.&rdquo

You have genuine regulation.
If the government can only say:

&ldquo Please reconsider.&rdquo

Then it is largely political theatre.

3. Does federal pre-emption survive?

There is another fascinating angle: China

This is where I think the article could go even further.
The accelerationists' strongest argument isn' t really:

&ldquo Regulation is bad.&rdquo

It' s:

&ldquo The United States cannot afford to slow down while China is racing ahead.&rdquo

That argument is extremely powerful inside Trump' s worldview.
But the Pentagon can turn it around:

&ldquo Precisely because China is racing ahead, we need to know what these systems can do before they are deployed.&rdquo

So the ultimate Trump policy may become:

Low regulation for ordinary AI

but

Heavy controls for frontier AI with military significance.

That is probably the most politically sustainable compromise.

What this means for the AI companies

This is where I would pay particularly close attention.

OpenAI / Google / Meta

Anthropic is particularly interesting

Anthropic' s behaviour creates a strange political position.
It effectively demonstrated:

&ldquo We have developed something sufficiently dangerous that we don' t want to release it.&rdquo

That gives the safety camp credibility.
But it also demonstrates something the accelerationists fear:
The company itself can decide what is safe.
That creates the government question:

Should Anthropic decide whether Mythos is safe, or should the Pentagon decide?

That is ultimately what this fight is about.

My probability tree for Trump AI policy

If I were analysing this as an investor rather than a political journalist, I' d assign something roughly like this:

Outcome	My probability	Market implication
Light-touch federal framework	45%	Bullish for AI
Mandatory frontier-model testing	35%	Moderate impact favours big players
Strict licensing regime	10%	Negative for smaller AI companies
Major national-security restrictions	10%	Potential AI-sector shock

I would not expect Trump suddenly to embrace Biden-style broad AI regulation.
That would contradict the economic and geopolitical logic of his administration.
Much more likely is:

Deregulation for the broad AI economy + national-security controls around the most powerful frontier systems.

The bigger investment conclusion

&ldquo Which company has the best chatbot?&rdquo

and more about:

&ldquo Which companies can develop frontier AI while satisfying governments that their systems are safe enough to deploy?&rdquo

That potentially favours the largest, best-capitalised companies.

And this is why I would NOT interpret Trump' s indecision simply as weakness

There is actually a rational reason for the hesitation.
The technology is changing faster than the regulatory framework can adapt.
Trump began with:

Speed first.

Anthropic' s Mythos effectively forced Washington to confront:

&ldquo What if speed itself becomes the security problem?&rdquo

chartiskao ( Date: 12-Aug-2026 06:45) Posted:

Core SGX Tech & Precision Engineering Beneficiaries

                             SGX Semiconductor Value-Chain
                                           │ 
         ┌ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┼ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┐ 
         ▼                                  ▼                                  ▼ 
┌ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┐        ┌ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┐        ┌ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┐ 
│      UMS Integration     │        │        AEM Holdings      │        │       Frencken Group     │ 
│    (SGX: 558 / Pure-Play │        │    (SGX: AWX / Advanced  │        │   (SGX: E28 / Diversified│ 
│    Deposition &  Etch)    │        │     Testing &  Packaging) │        │    High-NA EUV Mecha)    │ 
└ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┬ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┘        └ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┬ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┘        └ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┬ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┘ 
             │                                  │                                  │ 
             ▼                                  ▼                                  ▼ 
 Applied Materials Tier-1           Fabless AI / HPC Clients         ASML &  Healthcare / Industrial

1. Pure-Play Front-End Equipment: UMS Integration (SGX: 558)

Value Chain Exposure: UMS produces high-precision components and sub-assemblies for deposition, etch, and chemical mechanical planarization (CMP) tools used in semiconductor fabrication. Its revenue is tied to Applied Materials (AMAT).
AI Capex Catalyst: Unprecedented demand for advanced nodes and high-bandwidth memory (HBM) wafer fabrication equipment has driven strong order book visibility.
Financial & Valuation Profile: Strong balance sheet with a trailing net margin of ~17.4%. Offers attractive dividend yields (~4.4%), serving as a high-income, high-exposure proxy for global foundry equipment capex.

2. Advanced Packaging & AI Accelerators: AEM Holdings (SGX: AWX)

Value Chain Exposure: AEM provides system-level testing (SLT) solutions and automated handling equipment for advanced logic chips, graphics processing units (GPUs), and AI accelerators.
AI Capex Catalyst: As chiplet architectures and 2.5D/3D advanced packaging become standard for AI accelerators, testing complexity rises exponentially. AEM has benefited from a high-volume production ramp for its major fabless AI and high-performance computing (HPC) clients.
Financial & Valuation Profile: High operational beta stock. Q1 2026 revenue expanded 35.8% YoY to S$116.9M. Valuations reflect significant growth optionality as AEM expands testing partnerships beyond traditional lead customers into broader AI foundry environments.

3. Diversified Precision Mechatronics: Frencken Group (SGX: E28)

Value Chain Exposure: Outsourced design and high-precision manufacturing for extreme ultraviolet (EUV) lithography systems (supporting ASML), alongside medical, analytical, and industrial automation equipment.
AI Capex Catalyst: Demand for High-NA (Numerical Aperture) EUV lithography tooling is accelerating as chipmakers scale sub-2nm nodes and HBM architecture.
Financial & Valuation Profile: Offers downside protection relative to pure-play tech stocks due to non-semiconductor revenue streams (medical, industrial) that cushion macro volatility.

4. Specialized Niche & Component Plays

Grand Venture Technology (GVT) (SGX: JLB): Serves front-end semiconductor equipment, aerospace, and medical sectors. GVT is capturing increased market share as Tier-1 equipment makers expand production capacity in Southeast Asia.
Micro-Mechanics (Holdings) (SGX: 508): Manufactures consumable tools and high-precision parts used in semiconductor die-attach and wire-bonding processes. Benefits from high global fab utilization rates.

Key Financial & Operational Metrics Comparison

Stock Name & Ticker	Core AI/Tech Exposure Area	Revenue Growth Drivers	Trailing Net Margin	Dividend Profile	Risk Factor
UMS Integration (SGX: 558)	Deposition & Etch Tool Components	Applied Materials order ramp new Penang factory expansion.	~17.4%	4.0% &ndash 4.5% (Quarterly payouts)	Customer concentration in Applied Materials.
AEM Holdings (SGX: AWX)	System-Level Testing (SLT) for AI Chips	High-Performance Computing (HPC) & AI accelerator testing.	~6.5%	< 1.0% (Prioritizing reinvestment)	Customer concentration high earnings volatility.
Frencken Group (SGX: E28)	EUV Lithography Components & Modules	High-NA EUV rollout 2H26 memory capex rebound.	~4.5%	~1.0% (Stable dividend record)	Margins impacted by non-tech sector drags.
Grand Venture Tech (SGX: JLB)	Ultra-Precision CNC & Sub-assemblies	Front-end fab equipment module outsourcing.	~5.5% &ndash 7.0%	~1.5%	Integration costs from acquired facilities.

Investment Themes & Catalyst Roadmap

1. Operational ASP Increases Lift Profit Margins

2. Shift from Multiple Expansion to Earnings Realization

3. Institutional Inflows & Catalysts

Downside Risks to Monitor

Hyperscaler Capex Digestion: If global cloud tech giants slow infrastructure growth in 2027 due to enterprise AI monetization challenges, upstream order cancellations could affect SGX precision component suppliers.
Customer Concentration: UMS (AMAT) and AEM (Intel/key AI fabless) carry high customer concentration risks. Order volatility from primary customers impacts quarterly earnings.
Geopolitical Supply Chain Shocks: Continued friction in the Middle East and US tariff actions pose supply chain risks for chemical feedstocks and international freight costs.

chartiskao ( Date: 12-Aug-2026 06:34) Posted:

Executive Strategic Report: Singapore Macroeconomic Outlook & The Global AI Investment Cycle (2026)

Executive Summary

Singapore&rsquo s Ministry of Trade and Industry (MTI) upgraded the nation&rsquo s full-year 2026 Gross Domestic Product (GDP) growth forecast from 2.0%&ndash 4.0% to 4.5%&ndash 5.5%. This upward revision reflects a strong first half (H1 2026 GDP grew 6.1% YoY) and an accelerating global Artificial Intelligence (AI) capital expenditure cycle.
Private-sector economists have broadly revised their full-year projections upward, with Nomura issuing a high-end forecast of 5.7%&mdash exceeding the government' s official range. While the economic surge is primarily powered by manufacturing, wholesale trade, and financial services linked to advanced semiconductor hardware, it also exposes structural concentration risks. The sustainability of this growth trajectory depends on enterprise AI monetization and resilience against elevated geopolitical and energy cost head winds.

Macroeconomic Overview & Institutional Forecasts

Singapore&rsquo s Q2 2026 GDP expanded by 5.9% year-on-year, slightly moderating from Q1' s 6.3% pace but beating market consensus (5.8%). On a quarter-on-quarter seasonally adjusted basis, momentum remained solid at 1.4%.

       Singapore 2026 Full-Year GDP Forecast Growth Range
┌ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┐ 
│  MTI Upgraded Target: 4.5% - 5.5%                            │ 
└ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┘ 
  RHB          DBS / UOB       Maybank       MTI Top       Nomura
 [4.5%]         [5.0%]         [5.2%]        [5.5%]        [5.7%]
---|--------------|--------------|-------------|-------------|--->  GDP Growth

Institutional GDP Growth Forecast Revisions (2026)

Institution / Entity	Previous Forecast	Revised Forecast	Key Strategic Driver / Thesis
MTI (Official)	2.0% &ndash 4.0%	4.5% &ndash 5.5%	Robust H1 performance accelerated global AI capex.
Nomura	4.6%	5.7%	Bullish H2 projection (5.3%) multiple growth engines including Micron' s S$9.5B HBM plant.
Maybank	4.8%	5.2%	Triple tailwinds: AI capex boom, domestic construction, and safe-haven capital inflows.
DBS	4.3%	5.0%	Persistence of the global AI cycle despite geopolitical high base effects.
UOB	4.8%	5.0%	AI capex support, though semiconductor momentum may moderate in late H2.
RHB	4.5%	4.5%	Conservative outlier: expects H2 growth to cool to ~3.2% YoY due to narrow base.

Primary Sectoral Growth Engines

Singapore&rsquo s H1 2026 economic acceleration was driven by three core pillars:

Manufacturing Output (+12.5% YoY in Q2):
- Semiconductors & Precision Engineering: Expansion was fueled by global demand for memory chips (e.g., High-Bandwidth Memory / HBM), networking chips, and specialized wafer-fab equipment.
- Capacity Expansion: Projects like Micron' s S$9.5 billion High-Bandwidth Memory plant in Singapore provided structural capacity additions to meet global demand.
Wholesale Trade & Equipment:
- Global AI infrastructure deployment boosted the machinery, equipment, and supplies trade segments, positioning Singapore as a regional intermediate supply node.
Finance & Insurance (+6.2% YoY in Q2):
- Sustained credit expansion, strong investment banking fee generation, and double-digit growth in asset/fund management&mdash partially propelled by safe-haven capital inflows into Singapore.

Regional Footprint & Tech Ecosystem Synchronization

The global AI hardware build-out has triggered an export-led surge across key East Asian and Southeast Asian economies integrated into the technology supply chain.

                  ┌ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┐ 
                  │  Global AI Capital Expenditure  │ 
                  └ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┬ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┘ 
                                 │ 
         ┌ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┼ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┐ 
         ▼                        ▼                        ▼ 
┌ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┐    ┌ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┐    ┌ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┐ 
│  Taiwan (+9.64%)   │    │  S. Korea (+2.6%)  │    │  ASEAN Hubs        │ 
│  High-end Foundry  │    │  Memory / DRAM/HBM │    │  SG, MY, VN        │ 
└ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┘    └ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┘    └ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┘

Taiwan: GDP growth forecast upgraded to 9.64% (up from 7.71%) as AI server and logic chip manufacturing reached historic highs.
South Korea: Forecast raised to 2.6% (up from 2.0%) driven by high-bandwidth memory (HBM) and advanced DRAM exports.
Malaysia: Advance estimates indicate Q2 acceleration to 5.8%, marking its fastest manufacturing growth rate in nearly two decades due to semiconductor packaging and testing demand.
Vietnam: GDP expanded by 8.4%, leveraging assembly and electronics export hubs to satisfy AI hardware demand.
Divergent Economies: Economies driven primarily by domestic consumption or struggling with investment contraction&mdash such as the Philippines (downgraded to 3.5%&ndash 4.5%) and Indonesia (5.3% growth)&mdash did not participate equally in the AI export wave.

Strategic Vulnerabilities & Downside Risks

While short-term growth metrics are strong, economists highlight three critical structural risks:

1. Concentration Risk & Lack of Broad-Based Breadth

Growth remains asymmetric. As noted by analysts at RHB, manufacturing and associated wholesale trade account for the majority of GDP expansion. In contrast, domestic-facing sectors remain sluggish for instance, Food & Beverage (F& B) services contracted in Q2 due to sustained outbound travel by locals and subdued tourist spending.

2. The AI Capex Monetization Wall

A core macroeconomic risk is a potential sharp correction in global AI capital expenditures. Current hardware investments rely on expectations of long-term enterprise AI adoption. If end-user applications fail to monetize at scale, tech hyperscalers could reduce infrastructure spending, creating a drop in semiconductor demand.

3. Geopolitical & Energy Shock Spillover

Strait of Hormuz Dynamics: Disruption in energy transit routes has kept intermediate input and oil prices elevated.
Trade Policy Volatility: Potential additional foreign tariff escalations and trade frictions pose ongoing risks to Singapore' s high trade-to-GDP ratio.

Strategic Recommendations for Decision-Makers

Capitalize on H2 Hardware Demand: Enterprises in precision engineering and logistics should maximize throughput in H2 2026 while global order books for AI networking and memory chips remain elevated.
Hedge Against AI Hype Cycle Normalization: Corporate treasurers and strategists should prepare for growth to normalize closer to long-term trends (~3.0%&ndash 3.3%) in 2027 as high base effects set in and hyperscaler capex growth stabilizes.
Diversify Revenue Beyond Electronics: Businesses reliant on domestic consumer spend (F& B, retail) must adapt to changing local travel patterns and softer tourism spend by optimizing operational footprints and targeting localized value segments.

Useful To Me Not Useful To Me

chartiskao
Supreme

12-Aug-2026 06:45

chartiskao About Contact Quote this Post!

x 0
x 0

Alert Admin

Core SGX Tech & Precision Engineering Beneficiaries

                             SGX Semiconductor Value-Chain
                                           │ 
         ┌ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┼ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┐ 
         ▼                                  ▼                                  ▼ 
┌ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┐        ┌ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┐        ┌ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┐ 
│      UMS Integration     │        │        AEM Holdings      │        │       Frencken Group     │ 
│    (SGX: 558 / Pure-Play │        │    (SGX: AWX / Advanced  │        │   (SGX: E28 / Diversified│ 
│    Deposition &  Etch)    │        │     Testing &  Packaging) │        │    High-NA EUV Mecha)    │ 
└ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┬ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┘        └ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┬ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┘        └ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┬ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┘ 
             │                                  │                                  │ 
             ▼                                  ▼                                  ▼ 
 Applied Materials Tier-1           Fabless AI / HPC Clients         ASML &  Healthcare / Industrial

1. Pure-Play Front-End Equipment: UMS Integration (SGX: 558)

Value Chain Exposure: UMS produces high-precision components and sub-assemblies for deposition, etch, and chemical mechanical planarization (CMP) tools used in semiconductor fabrication. Its revenue is tied to Applied Materials (AMAT).
AI Capex Catalyst: Unprecedented demand for advanced nodes and high-bandwidth memory (HBM) wafer fabrication equipment has driven strong order book visibility.
Financial & Valuation Profile: Strong balance sheet with a trailing net margin of ~17.4%. Offers attractive dividend yields (~4.4%), serving as a high-income, high-exposure proxy for global foundry equipment capex.

2. Advanced Packaging & AI Accelerators: AEM Holdings (SGX: AWX)

Value Chain Exposure: AEM provides system-level testing (SLT) solutions and automated handling equipment for advanced logic chips, graphics processing units (GPUs), and AI accelerators.
AI Capex Catalyst: As chiplet architectures and 2.5D/3D advanced packaging become standard for AI accelerators, testing complexity rises exponentially. AEM has benefited from a high-volume production ramp for its major fabless AI and high-performance computing (HPC) clients.
Financial & Valuation Profile: High operational beta stock. Q1 2026 revenue expanded 35.8% YoY to S$116.9M. Valuations reflect significant growth optionality as AEM expands testing partnerships beyond traditional lead customers into broader AI foundry environments.

3. Diversified Precision Mechatronics: Frencken Group (SGX: E28)

Value Chain Exposure: Outsourced design and high-precision manufacturing for extreme ultraviolet (EUV) lithography systems (supporting ASML), alongside medical, analytical, and industrial automation equipment.
AI Capex Catalyst: Demand for High-NA (Numerical Aperture) EUV lithography tooling is accelerating as chipmakers scale sub-2nm nodes and HBM architecture.
Financial & Valuation Profile: Offers downside protection relative to pure-play tech stocks due to non-semiconductor revenue streams (medical, industrial) that cushion macro volatility.

4. Specialized Niche & Component Plays

Grand Venture Technology (GVT) (SGX: JLB): Serves front-end semiconductor equipment, aerospace, and medical sectors. GVT is capturing increased market share as Tier-1 equipment makers expand production capacity in Southeast Asia.
Micro-Mechanics (Holdings) (SGX: 508): Manufactures consumable tools and high-precision parts used in semiconductor die-attach and wire-bonding processes. Benefits from high global fab utilization rates.

Key Financial & Operational Metrics Comparison

Stock Name & Ticker	Core AI/Tech Exposure Area	Revenue Growth Drivers	Trailing Net Margin	Dividend Profile	Risk Factor
UMS Integration (SGX: 558)	Deposition & Etch Tool Components	Applied Materials order ramp new Penang factory expansion.	~17.4%	4.0% &ndash 4.5% (Quarterly payouts)	Customer concentration in Applied Materials.
AEM Holdings (SGX: AWX)	System-Level Testing (SLT) for AI Chips	High-Performance Computing (HPC) & AI accelerator testing.	~6.5%	< 1.0% (Prioritizing reinvestment)	Customer concentration high earnings volatility.
Frencken Group (SGX: E28)	EUV Lithography Components & Modules	High-NA EUV rollout 2H26 memory capex rebound.	~4.5%	~1.0% (Stable dividend record)	Margins impacted by non-tech sector drags.
Grand Venture Tech (SGX: JLB)	Ultra-Precision CNC & Sub-assemblies	Front-end fab equipment module outsourcing.	~5.5% &ndash 7.0%	~1.5%	Integration costs from acquired facilities.

Investment Themes & Catalyst Roadmap

1. Operational ASP Increases Lift Profit Margins

2. Shift from Multiple Expansion to Earnings Realization

3. Institutional Inflows & Catalysts

Downside Risks to Monitor

Hyperscaler Capex Digestion: If global cloud tech giants slow infrastructure growth in 2027 due to enterprise AI monetization challenges, upstream order cancellations could affect SGX precision component suppliers.
Customer Concentration: UMS (AMAT) and AEM (Intel/key AI fabless) carry high customer concentration risks. Order volatility from primary customers impacts quarterly earnings.
Geopolitical Supply Chain Shocks: Continued friction in the Middle East and US tariff actions pose supply chain risks for chemical feedstocks and international freight costs.

chartiskao ( Date: 12-Aug-2026 06:34) Posted:

Executive Strategic Report: Singapore Macroeconomic Outlook & The Global AI Investment Cycle (2026)

Executive Summary

Macroeconomic Overview & Institutional Forecasts

       Singapore 2026 Full-Year GDP Forecast Growth Range
┌ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┐ 
│  MTI Upgraded Target: 4.5% - 5.5%                            │ 
└ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┘ 
  RHB          DBS / UOB       Maybank       MTI Top       Nomura
 [4.5%]         [5.0%]         [5.2%]        [5.5%]        [5.7%]
---|--------------|--------------|-------------|-------------|--->  GDP Growth

Institutional GDP Growth Forecast Revisions (2026)

Institution / Entity	Previous Forecast	Revised Forecast	Key Strategic Driver / Thesis
MTI (Official)	2.0% &ndash 4.0%	4.5% &ndash 5.5%	Robust H1 performance accelerated global AI capex.
Nomura	4.6%	5.7%	Bullish H2 projection (5.3%) multiple growth engines including Micron' s S$9.5B HBM plant.
Maybank	4.8%	5.2%	Triple tailwinds: AI capex boom, domestic construction, and safe-haven capital inflows.
DBS	4.3%	5.0%	Persistence of the global AI cycle despite geopolitical high base effects.
UOB	4.8%	5.0%	AI capex support, though semiconductor momentum may moderate in late H2.
RHB	4.5%	4.5%	Conservative outlier: expects H2 growth to cool to ~3.2% YoY due to narrow base.

Primary Sectoral Growth Engines

Singapore&rsquo s H1 2026 economic acceleration was driven by three core pillars:

Manufacturing Output (+12.5% YoY in Q2):
- Semiconductors & Precision Engineering: Expansion was fueled by global demand for memory chips (e.g., High-Bandwidth Memory / HBM), networking chips, and specialized wafer-fab equipment.
- Capacity Expansion: Projects like Micron' s S$9.5 billion High-Bandwidth Memory plant in Singapore provided structural capacity additions to meet global demand.
Wholesale Trade & Equipment:
- Global AI infrastructure deployment boosted the machinery, equipment, and supplies trade segments, positioning Singapore as a regional intermediate supply node.
Finance & Insurance (+6.2% YoY in Q2):
- Sustained credit expansion, strong investment banking fee generation, and double-digit growth in asset/fund management&mdash partially propelled by safe-haven capital inflows into Singapore.

Regional Footprint & Tech Ecosystem Synchronization

The global AI hardware build-out has triggered an export-led surge across key East Asian and Southeast Asian economies integrated into the technology supply chain.

                  ┌ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┐ 
                  │  Global AI Capital Expenditure  │ 
                  └ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┬ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┘ 
                                 │ 
         ┌ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┼ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┐ 
         ▼                        ▼                        ▼ 
┌ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┐    ┌ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┐    ┌ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┐ 
│  Taiwan (+9.64%)   │    │  S. Korea (+2.6%)  │    │  ASEAN Hubs        │ 
│  High-end Foundry  │    │  Memory / DRAM/HBM │    │  SG, MY, VN        │ 
└ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┘    └ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┘    └ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┘

Taiwan: GDP growth forecast upgraded to 9.64% (up from 7.71%) as AI server and logic chip manufacturing reached historic highs.
South Korea: Forecast raised to 2.6% (up from 2.0%) driven by high-bandwidth memory (HBM) and advanced DRAM exports.
Malaysia: Advance estimates indicate Q2 acceleration to 5.8%, marking its fastest manufacturing growth rate in nearly two decades due to semiconductor packaging and testing demand.
Vietnam: GDP expanded by 8.4%, leveraging assembly and electronics export hubs to satisfy AI hardware demand.
Divergent Economies: Economies driven primarily by domestic consumption or struggling with investment contraction&mdash such as the Philippines (downgraded to 3.5%&ndash 4.5%) and Indonesia (5.3% growth)&mdash did not participate equally in the AI export wave.

Strategic Vulnerabilities & Downside Risks

While short-term growth metrics are strong, economists highlight three critical structural risks:

1. Concentration Risk & Lack of Broad-Based Breadth

2. The AI Capex Monetization Wall

3. Geopolitical & Energy Shock Spillover

Strait of Hormuz Dynamics: Disruption in energy transit routes has kept intermediate input and oil prices elevated.
Trade Policy Volatility: Potential additional foreign tariff escalations and trade frictions pose ongoing risks to Singapore' s high trade-to-GDP ratio.

Strategic Recommendations for Decision-Makers

Capitalize on H2 Hardware Demand: Enterprises in precision engineering and logistics should maximize throughput in H2 2026 while global order books for AI networking and memory chips remain elevated.
Hedge Against AI Hype Cycle Normalization: Corporate treasurers and strategists should prepare for growth to normalize closer to long-term trends (~3.0%&ndash 3.3%) in 2027 as high base effects set in and hyperscaler capex growth stabilizes.
Diversify Revenue Beyond Electronics: Businesses reliant on domestic consumer spend (F& B, retail) must adapt to changing local travel patterns and softer tourism spend by optimizing operational footprints and targeting localized value segments.

chartiskao ( Date: 11-Aug-2026 15:25) Posted:

&ldquo All I Have to Do Is Dream&rdquo is an even stronger metaphor for the current Asian AI fever than &ldquo Dream Lover.&rdquo
I can apply the song' s themes, but I can' t reproduce or adapt its copyrighted lyrics line-by-line. The investment metaphor can be expressed like this:

🌏 &ldquo All I Have to Do Is Dream&rdquo &mdash the Asian AI fever

The song is fundamentally about wanting something so strongly that imagination temporarily makes it feel within reach.
That describes today' s AI enthusiasm across Asia remarkably well.
Investors see a future where AI transforms:

🇸 🇬 Singapore &mdash data centres, cloud infrastructure, finance, tourism and digital services
🇹 🇼 Taiwan &mdash semiconductors, advanced packaging and AI hardware
🇰 🇷 South Korea &mdash memory chips, electronics and AI infrastructure
🇯 🇵 Japan &mdash robotics, automation and industrial AI
🇨 🇳 China &mdash domestic AI models, smartphones, EVs, robotics and industrial automation
🇭 🇰 Hong Kong &mdash financial technology and China' s technology capital markets

And the investment narrative becomes:
AI &rarr enormous future demand &rarr enormous future profits &rarr buy AI-related assets today.
That' s the dream.

But this is where the value investor wakes up

The important question is not whether the AI dream is real.
It is:

How much of the dream is already embedded in today' s share price?

A company can be an extraordinary AI beneficiary and still be a bad investment at an excessive valuation.
That' s the crucial distinction between AI technology and AI investment returns.
For a Buffett/Li Ka-shing-style investor, I would divide Asia' s AI opportunity into three layers:

Layer	What investors are dreaming about	What to examine
AI dream	Huge future AI economy	Is the opportunity actually achievable?
AI infrastructure	Chips, power, data centres, networks	Capital expenditure, debt, cash flow
AI cash machines	Companies actually earning money from AI	ROIC, FCF, margins, valuation

The biggest danger: paying for tomorrow twice

Suppose the market believes:
AI will transform Asia &rarr earnings will explode &rarr therefore today' s high valuation is justified.
The problem occurs when expectations become so high that even excellent AI growth isn' t enough to produce excellent investment returns.
That' s when the song' s dreamy quality becomes an investment warning.
The investor starts with:
&ldquo I can imagine what AI will become.&rdquo
But Buffett' s question is closer to:
&ldquo What am I actually buying today, and what cash will it generate?&rdquo

And this is why your dry-powder strategy matters

If the AI fever eventually produces a major correction, the opportunity may not necessarily be to abandon AI.
It could be to buy the second-order beneficiaries that become irrationally cheap.
For example:
AI boom &rarr massive data-centre construction &rarr electricity demand &rarr infrastructure investment &rarr financing requirements &rarr property/infrastructure beneficiaries &rarr banks and financial companies.
That is much more interesting from a value-investing perspective than simply chasing whichever AI stock is currently fashionable.
So I would describe the 2026 Asian AI market this way:

&ldquo All I Have to Do Is Dream&rdquo captures the psychology of the AI boom: investors can see an extraordinary future simply by imagining it. But the great value investor waits for the moment when the dream becomes mispriced reality.

And that distinction is particularly important in Singapore, because Singapore can benefit enormously from AI infrastructure without necessarily needing to own the most expensive AI stocks. Data centres, electricity, connectivity, financial services, property, tourism and infrastructure can all participate in the AI economy.
The dream creates the demand.
The infrastructure captures the spending.
The cash flow determines the investment.
The valuation determines the return.
https://www.youtube.com/watch?v=tbU3zdAgiX8& list=RDwVHAQX5sSaU& index=2

Useful To Me Not Useful To Me

chartiskao
Supreme

12-Aug-2026 06:34

chartiskao About Contact Quote this Post!

x 0
x 0

Alert Admin

Executive Strategic Report: Singapore Macroeconomic Outlook & The Global AI Investment Cycle (2026)

Executive Summary

Macroeconomic Overview & Institutional Forecasts

       Singapore 2026 Full-Year GDP Forecast Growth Range
┌ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┐ 
│  MTI Upgraded Target: 4.5% - 5.5%                            │ 
└ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┘ 
  RHB          DBS / UOB       Maybank       MTI Top       Nomura
 [4.5%]         [5.0%]         [5.2%]        [5.5%]        [5.7%]
---|--------------|--------------|-------------|-------------|--->  GDP Growth

Institutional GDP Growth Forecast Revisions (2026)

Institution / Entity	Previous Forecast	Revised Forecast	Key Strategic Driver / Thesis
MTI (Official)	2.0% &ndash 4.0%	4.5% &ndash 5.5%	Robust H1 performance accelerated global AI capex.
Nomura	4.6%	5.7%	Bullish H2 projection (5.3%) multiple growth engines including Micron' s S$9.5B HBM plant.
Maybank	4.8%	5.2%	Triple tailwinds: AI capex boom, domestic construction, and safe-haven capital inflows.
DBS	4.3%	5.0%	Persistence of the global AI cycle despite geopolitical high base effects.
UOB	4.8%	5.0%	AI capex support, though semiconductor momentum may moderate in late H2.
RHB	4.5%	4.5%	Conservative outlier: expects H2 growth to cool to ~3.2% YoY due to narrow base.

Primary Sectoral Growth Engines

Singapore&rsquo s H1 2026 economic acceleration was driven by three core pillars:

Manufacturing Output (+12.5% YoY in Q2):
- Semiconductors & Precision Engineering: Expansion was fueled by global demand for memory chips (e.g., High-Bandwidth Memory / HBM), networking chips, and specialized wafer-fab equipment.
- Capacity Expansion: Projects like Micron' s S$9.5 billion High-Bandwidth Memory plant in Singapore provided structural capacity additions to meet global demand.
Wholesale Trade & Equipment:
- Global AI infrastructure deployment boosted the machinery, equipment, and supplies trade segments, positioning Singapore as a regional intermediate supply node.
Finance & Insurance (+6.2% YoY in Q2):
- Sustained credit expansion, strong investment banking fee generation, and double-digit growth in asset/fund management&mdash partially propelled by safe-haven capital inflows into Singapore.

Regional Footprint & Tech Ecosystem Synchronization

The global AI hardware build-out has triggered an export-led surge across key East Asian and Southeast Asian economies integrated into the technology supply chain.

                  ┌ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┐ 
                  │  Global AI Capital Expenditure  │ 
                  └ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┬ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┘ 
                                 │ 
         ┌ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┼ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┐ 
         ▼                        ▼                        ▼ 
┌ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┐    ┌ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┐    ┌ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┐ 
│  Taiwan (+9.64%)   │    │  S. Korea (+2.6%)  │    │  ASEAN Hubs        │ 
│  High-end Foundry  │    │  Memory / DRAM/HBM │    │  SG, MY, VN        │ 
└ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┘    └ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┘    └ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ┘

Taiwan: GDP growth forecast upgraded to 9.64% (up from 7.71%) as AI server and logic chip manufacturing reached historic highs.
South Korea: Forecast raised to 2.6% (up from 2.0%) driven by high-bandwidth memory (HBM) and advanced DRAM exports.
Malaysia: Advance estimates indicate Q2 acceleration to 5.8%, marking its fastest manufacturing growth rate in nearly two decades due to semiconductor packaging and testing demand.
Vietnam: GDP expanded by 8.4%, leveraging assembly and electronics export hubs to satisfy AI hardware demand.
Divergent Economies: Economies driven primarily by domestic consumption or struggling with investment contraction&mdash such as the Philippines (downgraded to 3.5%&ndash 4.5%) and Indonesia (5.3% growth)&mdash did not participate equally in the AI export wave.

Strategic Vulnerabilities & Downside Risks

While short-term growth metrics are strong, economists highlight three critical structural risks:

1. Concentration Risk & Lack of Broad-Based Breadth

2. The AI Capex Monetization Wall

3. Geopolitical & Energy Shock Spillover

Strait of Hormuz Dynamics: Disruption in energy transit routes has kept intermediate input and oil prices elevated.
Trade Policy Volatility: Potential additional foreign tariff escalations and trade frictions pose ongoing risks to Singapore' s high trade-to-GDP ratio.

Strategic Recommendations for Decision-Makers

Capitalize on H2 Hardware Demand: Enterprises in precision engineering and logistics should maximize throughput in H2 2026 while global order books for AI networking and memory chips remain elevated.
Hedge Against AI Hype Cycle Normalization: Corporate treasurers and strategists should prepare for growth to normalize closer to long-term trends (~3.0%&ndash 3.3%) in 2027 as high base effects set in and hyperscaler capex growth stabilizes.
Diversify Revenue Beyond Electronics: Businesses reliant on domestic consumer spend (F& B, retail) must adapt to changing local travel patterns and softer tourism spend by optimizing operational footprints and targeting localized value segments.

chartiskao ( Date: 11-Aug-2026 15:25) Posted:

🌏 &ldquo All I Have to Do Is Dream&rdquo &mdash the Asian AI fever

🇸 🇬 Singapore &mdash data centres, cloud infrastructure, finance, tourism and digital services
🇹 🇼 Taiwan &mdash semiconductors, advanced packaging and AI hardware
🇰 🇷 South Korea &mdash memory chips, electronics and AI infrastructure
🇯 🇵 Japan &mdash robotics, automation and industrial AI
🇨 🇳 China &mdash domestic AI models, smartphones, EVs, robotics and industrial automation
🇭 🇰 Hong Kong &mdash financial technology and China' s technology capital markets

And the investment narrative becomes:
AI &rarr enormous future demand &rarr enormous future profits &rarr buy AI-related assets today.
That' s the dream.

But this is where the value investor wakes up

The important question is not whether the AI dream is real.
It is:

How much of the dream is already embedded in today' s share price?

Layer	What investors are dreaming about	What to examine
AI dream	Huge future AI economy	Is the opportunity actually achievable?
AI infrastructure	Chips, power, data centres, networks	Capital expenditure, debt, cash flow
AI cash machines	Companies actually earning money from AI	ROIC, FCF, margins, valuation

The biggest danger: paying for tomorrow twice

And this is why your dry-powder strategy matters

&ldquo All I Have to Do Is Dream&rdquo captures the psychology of the AI boom: investors can see an extraordinary future simply by imagining it. But the great value investor waits for the moment when the dream becomes mispriced reality.

chartiskao ( Date: 11-Aug-2026 15:22) Posted:

Bobby Darin&rsquo s &ldquo Dream Lover&rdquo captures the idea of longing for an idealized dream. Applied to the 2026 AI market, the metaphor would be:

The market has fallen in love with its &ldquo AI Dream Lover&rdquo &mdash but investors now have to distinguish the dream from the cash flows.

The parallel is particularly interesting after the recent AI-driven market enthusiasm:

The dream: AI will transform productivity, software, chips, data centres and virtually every industry.
The excitement: Investors bid up companies connected to the AI theme, sometimes far ahead of present earnings.
The reality check: Enormous capital expenditure, electricity demand, chip costs and competition eventually have to produce measurable returns.
The contrarian question: Which companies are actually generating durable free cash flow from AI rather than simply selling the &ldquo dream&rdquo ?
The Buffett/Li Ka-shing approach: Don' t chase the dream simply because everyone believes in it. Wait for a price where the underlying business provides a margin of safety.

So your phrase &ldquo Dream Lover tops global AI fever&rdquo could work as a memorable investment metaphor:

🎵 From &ldquo Dream Lover&rdquo to &ldquo AI Dream Lover&rdquo

The late-1950s song was about waiting for the ideal dream partner to appear. In today' s market, AI has arguably become the global investor' s Dream Lover: exciting, transformative and capable of inspiring enormous expectations.
But eventually, just as a dream has to meet reality when you wake up, AI valuations have to meet earnings, cash flow and returns on invested capital.
For a value investor, the key question isn' t:
&ldquo Will AI change the world?&rdquo
It probably will.
The more important question is:
&ldquo At today' s price, how much of that future has already been paid for?&rdquo
That is where a Buffett-style investor becomes especially cautious&mdash and where keeping dry powder can become valuable if the AI fever produces a broad market correction.

https://www.youtube.com/watch?v=wVHAQX5sSaU& list=RDwVHAQX5sSaU& start_radio=1

Useful To Me Not Useful To Me

chartiskao
Supreme

11-Aug-2026 15:25

chartiskao About Contact Quote this Post!

x 0
x 0

Alert Admin

🌏 &ldquo All I Have to Do Is Dream&rdquo &mdash the Asian AI fever

🇸 🇬 Singapore &mdash data centres, cloud infrastructure, finance, tourism and digital services
🇹 🇼 Taiwan &mdash semiconductors, advanced packaging and AI hardware
🇰 🇷 South Korea &mdash memory chips, electronics and AI infrastructure
🇯 🇵 Japan &mdash robotics, automation and industrial AI
🇨 🇳 China &mdash domestic AI models, smartphones, EVs, robotics and industrial automation
🇭 🇰 Hong Kong &mdash financial technology and China' s technology capital markets

And the investment narrative becomes:
AI &rarr enormous future demand &rarr enormous future profits &rarr buy AI-related assets today.
That' s the dream.

But this is where the value investor wakes up

The important question is not whether the AI dream is real.
It is:

How much of the dream is already embedded in today' s share price?

Layer	What investors are dreaming about	What to examine
AI dream	Huge future AI economy	Is the opportunity actually achievable?
AI infrastructure	Chips, power, data centres, networks	Capital expenditure, debt, cash flow
AI cash machines	Companies actually earning money from AI	ROIC, FCF, margins, valuation

The biggest danger: paying for tomorrow twice

And this is why your dry-powder strategy matters

&ldquo All I Have to Do Is Dream&rdquo captures the psychology of the AI boom: investors can see an extraordinary future simply by imagining it. But the great value investor waits for the moment when the dream becomes mispriced reality.

chartiskao ( Date: 11-Aug-2026 15:22) Posted:

Bobby Darin&rsquo s &ldquo Dream Lover&rdquo captures the idea of longing for an idealized dream. Applied to the 2026 AI market, the metaphor would be:

The market has fallen in love with its &ldquo AI Dream Lover&rdquo &mdash but investors now have to distinguish the dream from the cash flows.

The parallel is particularly interesting after the recent AI-driven market enthusiasm:

The dream: AI will transform productivity, software, chips, data centres and virtually every industry.
The excitement: Investors bid up companies connected to the AI theme, sometimes far ahead of present earnings.
The reality check: Enormous capital expenditure, electricity demand, chip costs and competition eventually have to produce measurable returns.
The contrarian question: Which companies are actually generating durable free cash flow from AI rather than simply selling the &ldquo dream&rdquo ?
The Buffett/Li Ka-shing approach: Don' t chase the dream simply because everyone believes in it. Wait for a price where the underlying business provides a margin of safety.

So your phrase &ldquo Dream Lover tops global AI fever&rdquo could work as a memorable investment metaphor:

🎵 From &ldquo Dream Lover&rdquo to &ldquo AI Dream Lover&rdquo

https://www.youtube.com/watch?v=wVHAQX5sSaU& list=RDwVHAQX5sSaU& start_radio=1

chartistkaohz ( Date: 10-Aug-2026 14:02) Posted:

I traced the two stocks through their listing history and the major Hong Kong/property-market crises. One important caveat: very long-run Hong Kong share-price data are affected by splits, bonus issues and other corporate actions, so comparing an old nominal quote directly with today's HK$ price can be misleading. I therefore separate historical market prices from adjusted/total-return data where possible.
1. Starting point: the two companies
Company
Stock
Founded
Listed
IPO/listing price
Henderson Land Development
0012
1976
23 Jul 1981
Historical listing-price field is not reported by HKEX/ETNet
New World Development
0017
1970
23 Nov 1972
HK$2.00
ETNet confirms Henderson's 23 July 1981 listing date, although its database does not give a listing price. �
etnet 經濟通 +1
For New World, contemporary historical accounts state that it issued 96.75 million shares at HK$2 each when it went public in 1972. �
Encyclopedia.com +1
So New World gives us a clean IPO starting point of HK$2. Henderson requires more caution because I don't want to invent an IPO price that the historical exchange data do not establish.
2. The first great Hong Kong property cycle: 1970s?1980s
New World is particularly interesting because it went public just as Hong Kong was entering an extraordinary period of urban/property development.
1972 IPO: HK$2
The company used the IPO proceeds to help finance the New World Centre project in Tsim Sha Tsui. �
Encyclopedia.com
Then Hong Kong experienced:
1973?74 global/oil shock
1970s property volatility
1980s boom
1983 Hong Kong/China political uncertainty
1987 global stock-market crash.
This is the first lesson from New World's history:
Property developers can survive enormous share-price cycles because the underlying land bank can appreciate enormously over decades.
But shareholders still experience huge drawdowns along the way.
3. 1987 crash
The 1987 global stock-market crash was particularly severe for Hong Kong.
Property companies were highly sensitive to:
interest rates
stock-market liquidity
property transactions
investor confidence.
This is an important precedent for today's situation.
The share-price collapse does not necessarily mean the property assets have become worthless.
It often means:
investors suddenly demand a much higher risk premium for owning property equities.
That distinction becomes very important later.
4. 1997 Asian Financial Crisis
This was probably the most important historical comparison for your current thesis.
Hong Kong property had reached extraordinary valuations before the handover.
Then:
1997 Asian Financial Crisis
↓
capital outflow
↓
interest rates rise
↓
property prices collapse
↓
developers' share prices collapse
↓
NAV discounts widen.
This was a classic property + leverage + currency/interest-rate shock.
And Hong Kong property eventually recovered.
That history is one reason I don't automatically interpret today's depressed Henderson/New World prices as evidence that the underlying land assets have permanently lost their value.
5. Henderson's long-term pattern is particularly revealing
The available historical series from 2000 onward shows the enormous cyclicality.
Henderson's annual share-price performance included:
Year
Approx. price performance
2000
~0%
2001
− 9.9%
2002
− 24.6%
2003
+32.9%
2004
+20.3%
2005
− 10.0%
2006
+20.7%
2007
+64.9%
2008
− 59.5%
2009
+98.8%
2010
− 7.3%
2011
− 30.0%
2012
+49.1%
2013
− 12.5%
2014
+29.7%
2015
+0.6%
2016
− 4.0%
2017
+30.6%
2018
− 12.5%
2019
+5.5%
2020
− 18.4%
2021
+6.4%
2022
− 20.8%
2023
− 13.7%
2024
+8.6%
2025
+19.5%
These figures come from a historical price series and should be treated as adjusted historical performance rather than literal year-end HK$ quotes. �
Market Cap Of
Look at the pattern:
2008 − 59.5%
followed by
2009 +98.8%.
That's an extraordinary example of what happens when a property stock goes from:
crisis valuation → liquidity recovery → property recovery.
6. The 2008 Global Financial Crisis
This is perhaps the most useful historical precedent for what you're asking about.
Henderson:
2007: +64.9%
↓
2008: − 59.5%
↓
2009: +98.8%
The monthly historical series shows Henderson around:
HK$37.74 in Dec 2007
HK$34.30 in Jan 2008
HK$13.91 in Oct 2008
HK$14.73 in Dec 2008
HK$15.14 in Mar 2009
HK$29.97 in Dec 2009. �
Digrin
So the stock essentially went through:
boom → crash → almost 100% rebound
within a relatively short period.
That's exactly why property equities can be extremely powerful after major deratings.
7. New World was even more volatile
The New World historical series shows an even more dramatic cycle.
Before the financial crisis:
Period
Historical price indication
Dec 2006
~HK$53.46*
Dec 2007
~HK$94.74*
Jan 2008
~HK$80.57*
May 2008
~HK$66.91*
Oct 2008
~HK$21.30*
Dec 2008
~HK$26.83*
Mar 2009
~HK$26.39*
Dec 2009
~HK$54.48*
*These are the historical-series figures displayed by Digrin they include corporate-action adjustments and therefore should not be compared directly with today's unadjusted HK$6?7 quotation. �
Digrin
The important thing is the shape, not the absolute number:
2007 boom
→
2008 collapse
→
2009 recovery
8. Then came SARS in 2003?04
This is another useful precedent for New World.
New World reported a loss in 2004 during the SARS-related property downturn. The company later recovered. �
Financial Times
That matters because it demonstrates something fundamental:
A property developer can suffer a temporary collapse in earnings without its long-term property franchise being destroyed.
But today's New World situation is more complicated because the company has much greater debt pressure than Henderson.
9. 2015?2018: another property/China boom
Hong Kong property stocks benefited from:
mainland capital
low interest rates
China's growth
strong Hong Kong property prices.
Henderson's historical performance:
2017 +30.6%
before:
2018 − 12.5%. �
Market Cap Of
Again:
property stocks don't move smoothly with NAV.
They move around NAV.
10. 2019?2020: the beginning of the modern collapse
This is where today's cycle really begins.
Hong Kong faced:
2019 protests
↓
COVID in 2020
↓
three years of strict pandemic restrictions
↓
population outflow
↓
weak retail/office demand
↓
US Federal Reserve rate hikes
↓
Hong Kong rates rise because of the USD peg.
The FT documented residential prices falling roughly 25% from their 2021 peak and office rents falling nearly 40% from their 2019 peak during the subsequent downturn. �
Financial Times
11. 2020 COVID crash
Henderson:
2020 − 18.4%
according to the historical adjusted series. �
Market Cap Of
This is important because 2020 wasn't actually the bottom of the current Hong Kong property cycle.
Instead, the real damage came later.
12. 2021?2025: the structural property bear market
This was the really painful period.
You had:
China property crisis

Evergrande

Hong Kong property correction

US rate hikes

HKD/USD monetary constraint

zero-COVID

capital outflows

weak office demand

AI/technology investment boom.
Hong Kong developers were hit particularly hard.
New World became the extreme example.
Reuters reported in 2024 that New World's shares had fallen roughly 80% from mid-2021, while the company warned of a potential HK$20 billion loss. �
Reuters
The FT likewise described New World's share price as having fallen more than 80% since 2020. �
Financial Times
13. New World's 2024 crisis was not merely a normal property correction
This is where I would be careful with your rotation thesis.
New World wasn't simply:
"cheap because property is unpopular."
It became:
cheap because investors feared balance-sheet/liquidity risk.
In September 2024 the stock fell to HK$6.74, a 21-year low, after the company projected a potential HK$20 billion loss. �
Reuters
Then in 2025:
net gearing rose above 88%
debt became the central investment issue
the company deferred certain perpetual-bond coupon payments. �
Reuters +1
That makes New World fundamentally different from Henderson.
14. The 2025 refinancing was a major turning point
In July 2025, New World completed an HK$88.2 billion refinancing package.
The earliest maturity under the new structure was June 2028. �
Reuters
This didn't solve the entire balance-sheet problem.
But it changed the immediate question from:
"Could New World face a near-term liquidity crisis?"
to:
"Can management use the additional time to sell assets, generate cash and reduce leverage?"
That's a very important transition.
15. And now we arrive at 2026
This is where the historical comparison becomes fascinating.
Henderson
2026 peak:
HK$35.66
The share price subsequently fell significantly the historical 52-week range reported by ETNet was HK$17.80?HK$35.66 earlier in 2026. �
webuat01.etnet.com.hk
By early August, Henderson was around the high-HK$20s. Morningstar showed HK$26.88 on Aug. 7. �
Morningstar
So Henderson has gone from:
2026 peak ~HK$35.66
→
~HK$26.9
≈ − 25%.
16. New World 2026
New World reached approximately:
HK$12.45
during the 2026 recovery.
But it subsequently fell back toward the:
HK$6?7 range.
Google Finance currently shows a 52-week range of approximately HK$6.09?HK$12.45, with the stock around HK$6?7. �
Google
So:
HK$12.45 → ~HK$6.7
is roughly a:
46% decline
from its 2026 high.
This is a very different risk/reward setup from Henderson.
17. Put the whole history together
Henderson Land
1981 listing
↓
1987 crash
↓
1997 Asian Financial Crisis
↓
2000?03 technology/property downturn
↓
2008 Global Financial Crisis
↓
2009 massive recovery
↓
2011 correction
↓
2015?18 property cycle
↓
2019 Hong Kong crisis
↓
2020 COVID
↓
2021?25 property/rates bear market
↓
2026 recovery
↓
2026 correction
New World
1972 IPO HK$2
↓
1974 global/property downturn
↓
1987 crash
↓
1997 Asian Financial Crisis
↓
2003 SARS
↓
2008 Global Financial Crisis
↓
2009 recovery
↓
2015?18 China/HK property boom
↓
2019 Hong Kong crisis
↓
2020 COVID
↓
2021?25 China/HK property collapse
↓
2024 liquidity/balance-sheet crisis
↓
2025 refinancing
↓
2026 recovery
↓
2026 correction
18. The most important historical comparison
Look at these two episodes:
Henderson 2008?09
− 59.5%
↓
+98.8%
Current cycle
2021?25:
multi-year derating
↓
2026:
initial recovery
↓
2026:
correction
That doesn't guarantee another 2009-style rebound.
But it tells us something important:
The biggest returns in property stocks historically occur when investors buy during the transition from "survival" to "normalization."
Not when the property market is already obviously booming.
19. Henderson versus New World today
This is the most important conclusion from tracing their histories.

Henderson Land
New World
Listing
1981
1972
2026 high
~HK$35.66
~HK$12.45
Current Aug 2026
~HK$27
~HK$6?7
Fall from 2026 high
~25%
~46%
Balance sheet
Much stronger
Much weaker
NAV story
Strong
Potentially enormous
Liquidity risk
Lower
Much higher
Recovery upside
High
Very high
Probability of permanent impairment
Lower
Higher
Investment character
Value
Distressed value
20. Why this matters for your rotation thesis
This historical study actually strengthens the argument we were discussing earlier.
The 2020?25 property sell-off is not unprecedented.
Hong Kong property equities have repeatedly gone through:
boom → leverage → crisis → huge derating → stabilization → NAV recovery → re-rating.
The critical difference this time is that New World has a balance-sheet problem, whereas Henderson is much closer to a conventional property/NAV valuation opportunity.
Therefore I would think of the two stocks differently:
Henderson
2026?27 thesis:
Interest rates stabilize
→ Hong Kong property stabilizes
→ NAV discount narrows
→ developer earnings normalize
→ dividend remains attractive
→ share price re-rates.
New World
2026?28 thesis:
Refinancing buys time
→ asset sales
→ debt reduction
→ property market recovery
→ liquidity risk premium falls
→ NAV discount collapses.
The second scenario has much greater upside if everything works, but also a much greater possibility of permanent capital loss if deleveraging fails.
The historical lesson I would take
The most interesting thing isn't that Henderson and New World have fallen.
It is that they have fallen after an unusually long 2020?25 property/rates bear market, while the market is simultaneously extremely concentrated in AI, technology and banks.
That is precisely the environment in which a long-duration value rotation can eventually begin.
But I would not call the bottom solely from the share price.
For Henderson, I'd watch:
P/NAV + dividend yield + debt cost + Hong Kong property prices.
For New World, I'd put debt/refinancing and asset sales ahead of P/NAV.
And the historical data show why: Henderson's 2008?09 experience demonstrates how violently a quality property stock can rebound after a major derating, while New World's recent history demonstrates that a cheap-looking property stock can remain cheap when leverage becomes the dominant concern. �
Market Cap Of +1
If you want, the next useful step is to � ⁠ build a **1972/1981?2026 crisis-by-crisis table for both companies showing IPO price → crisis low → recovery high → percentage loss → percentage recovery → dividend received**, which would make the comparison much clearer.

Useful To Me Not Useful To Me

chartiskao
Supreme

11-Aug-2026 15:22

chartiskao About Contact Quote this Post!

x 0
x 0

Alert Admin

Bobby Darin&rsquo s &ldquo Dream Lover&rdquo captures the idea of longing for an idealized dream. Applied to the 2026 AI market, the metaphor would be:

The market has fallen in love with its &ldquo AI Dream Lover&rdquo &mdash but investors now have to distinguish the dream from the cash flows.

The parallel is particularly interesting after the recent AI-driven market enthusiasm:

The dream: AI will transform productivity, software, chips, data centres and virtually every industry.
The excitement: Investors bid up companies connected to the AI theme, sometimes far ahead of present earnings.
The reality check: Enormous capital expenditure, electricity demand, chip costs and competition eventually have to produce measurable returns.
The contrarian question: Which companies are actually generating durable free cash flow from AI rather than simply selling the &ldquo dream&rdquo ?
The Buffett/Li Ka-shing approach: Don' t chase the dream simply because everyone believes in it. Wait for a price where the underlying business provides a margin of safety.

So your phrase &ldquo Dream Lover tops global AI fever&rdquo could work as a memorable investment metaphor:

🎵 From &ldquo Dream Lover&rdquo to &ldquo AI Dream Lover&rdquo

https://www.youtube.com/watch?v=wVHAQX5sSaU& list=RDwVHAQX5sSaU& start_radio=1

chartistkaohz ( Date: 10-Aug-2026 14:02) Posted:

Useful To Me Not Useful To Me

chartistkaohz
Supreme

10-Aug-2026 14:02

chartistkaohz About Contact Quote this Post!

x 0
x 0

Alert Admin I traced the two stocks through their listing history and the major Hong Kong/property-market crises. One important caveat: very long-run Hong Kong share-price data are affected by splits, bonus issues and other corporate actions, so comparing an old nominal quote directly with today's HK$ price can be misleading. I therefore separate historical market prices from adjusted/total-return data where possible.
1. Starting point: the two companies
Company
Stock
Founded
Listed
IPO/listing price
Henderson Land Development
0012
1976
23 Jul 1981
Historical listing-price field is not reported by HKEX/ETNet
New World Development
0017
1970
23 Nov 1972
HK$2.00
ETNet confirms Henderson's 23 July 1981 listing date, although its database does not give a listing price. �
etnet 經濟通 +1
For New World, contemporary historical accounts state that it issued 96.75 million shares at HK$2 each when it went public in 1972. �
Encyclopedia.com +1
So New World gives us a clean IPO starting point of HK$2. Henderson requires more caution because I don't want to invent an IPO price that the historical exchange data do not establish.
2. The first great Hong Kong property cycle: 1970s?1980s
New World is particularly interesting because it went public just as Hong Kong was entering an extraordinary period of urban/property development.
1972 IPO: HK$2
The company used the IPO proceeds to help finance the New World Centre project in Tsim Sha Tsui. �
Encyclopedia.com
Then Hong Kong experienced:
1973?74 global/oil shock
1970s property volatility
1980s boom
1983 Hong Kong/China political uncertainty
1987 global stock-market crash.
This is the first lesson from New World's history:
Property developers can survive enormous share-price cycles because the underlying land bank can appreciate enormously over decades.
But shareholders still experience huge drawdowns along the way.
3. 1987 crash
The 1987 global stock-market crash was particularly severe for Hong Kong.
Property companies were highly sensitive to:
interest rates
stock-market liquidity
property transactions
investor confidence.
This is an important precedent for today's situation.
The share-price collapse does not necessarily mean the property assets have become worthless.
It often means:
investors suddenly demand a much higher risk premium for owning property equities.
That distinction becomes very important later.
4. 1997 Asian Financial Crisis
This was probably the most important historical comparison for your current thesis.
Hong Kong property had reached extraordinary valuations before the handover.
Then:
1997 Asian Financial Crisis
↓
capital outflow
↓
interest rates rise
↓
property prices collapse
↓
developers' share prices collapse
↓
NAV discounts widen.
This was a classic property + leverage + currency/interest-rate shock.
And Hong Kong property eventually recovered.
That history is one reason I don't automatically interpret today's depressed Henderson/New World prices as evidence that the underlying land assets have permanently lost their value.
5. Henderson's long-term pattern is particularly revealing
The available historical series from 2000 onward shows the enormous cyclicality.
Henderson's annual share-price performance included:
Year
Approx. price performance
2000
~0%
2001
− 9.9%
2002
− 24.6%
2003
+32.9%
2004
+20.3%
2005
− 10.0%
2006
+20.7%
2007
+64.9%
2008
− 59.5%
2009
+98.8%
2010
− 7.3%
2011
− 30.0%
2012
+49.1%
2013
− 12.5%
2014
+29.7%
2015
+0.6%
2016
− 4.0%
2017
+30.6%
2018
− 12.5%
2019
+5.5%
2020
− 18.4%
2021
+6.4%
2022
− 20.8%
2023
− 13.7%
2024
+8.6%
2025
+19.5%
These figures come from a historical price series and should be treated as adjusted historical performance rather than literal year-end HK$ quotes. �
Market Cap Of
Look at the pattern:
2008 − 59.5%
followed by
2009 +98.8%.
That's an extraordinary example of what happens when a property stock goes from:
crisis valuation → liquidity recovery → property recovery.
6. The 2008 Global Financial Crisis
This is perhaps the most useful historical precedent for what you're asking about.
Henderson:
2007: +64.9%
↓
2008: − 59.5%
↓
2009: +98.8%
The monthly historical series shows Henderson around:
HK$37.74 in Dec 2007
HK$34.30 in Jan 2008
HK$13.91 in Oct 2008
HK$14.73 in Dec 2008
HK$15.14 in Mar 2009
HK$29.97 in Dec 2009. �
Digrin
So the stock essentially went through:
boom → crash → almost 100% rebound
within a relatively short period.
That's exactly why property equities can be extremely powerful after major deratings.
7. New World was even more volatile
The New World historical series shows an even more dramatic cycle.
Before the financial crisis:
Period
Historical price indication
Dec 2006
~HK$53.46*
Dec 2007
~HK$94.74*
Jan 2008
~HK$80.57*
May 2008
~HK$66.91*
Oct 2008
~HK$21.30*
Dec 2008
~HK$26.83*
Mar 2009
~HK$26.39*
Dec 2009
~HK$54.48*
*These are the historical-series figures displayed by Digrin they include corporate-action adjustments and therefore should not be compared directly with today's unadjusted HK$6?7 quotation. �
Digrin
The important thing is the shape, not the absolute number:
2007 boom
→
2008 collapse
→
2009 recovery
8. Then came SARS in 2003?04
This is another useful precedent for New World.
New World reported a loss in 2004 during the SARS-related property downturn. The company later recovered. �
Financial Times
That matters because it demonstrates something fundamental:
A property developer can suffer a temporary collapse in earnings without its long-term property franchise being destroyed.
But today's New World situation is more complicated because the company has much greater debt pressure than Henderson.
9. 2015?2018: another property/China boom
Hong Kong property stocks benefited from:
mainland capital
low interest rates
China's growth
strong Hong Kong property prices.
Henderson's historical performance:
2017 +30.6%
before:
2018 − 12.5%. �
Market Cap Of
Again:
property stocks don't move smoothly with NAV.
They move around NAV.
10. 2019?2020: the beginning of the modern collapse
This is where today's cycle really begins.
Hong Kong faced:
2019 protests
↓
COVID in 2020
↓
three years of strict pandemic restrictions
↓
population outflow
↓
weak retail/office demand
↓
US Federal Reserve rate hikes
↓
Hong Kong rates rise because of the USD peg.
The FT documented residential prices falling roughly 25% from their 2021 peak and office rents falling nearly 40% from their 2019 peak during the subsequent downturn. �
Financial Times
11. 2020 COVID crash
Henderson:
2020 − 18.4%
according to the historical adjusted series. �
Market Cap Of
This is important because 2020 wasn't actually the bottom of the current Hong Kong property cycle.
Instead, the real damage came later.
12. 2021?2025: the structural property bear market
This was the really painful period.
You had:
China property crisis

Evergrande

Hong Kong property correction

US rate hikes

HKD/USD monetary constraint

zero-COVID

capital outflows

weak office demand

AI/technology investment boom.
Hong Kong developers were hit particularly hard.
New World became the extreme example.
Reuters reported in 2024 that New World's shares had fallen roughly 80% from mid-2021, while the company warned of a potential HK$20 billion loss. �
Reuters
The FT likewise described New World's share price as having fallen more than 80% since 2020. �
Financial Times
13. New World's 2024 crisis was not merely a normal property correction
This is where I would be careful with your rotation thesis.
New World wasn't simply:
"cheap because property is unpopular."
It became:
cheap because investors feared balance-sheet/liquidity risk.
In September 2024 the stock fell to HK$6.74, a 21-year low, after the company projected a potential HK$20 billion loss. �
Reuters
Then in 2025:
net gearing rose above 88%
debt became the central investment issue
the company deferred certain perpetual-bond coupon payments. �
Reuters +1
That makes New World fundamentally different from Henderson.
14. The 2025 refinancing was a major turning point
In July 2025, New World completed an HK$88.2 billion refinancing package.
The earliest maturity under the new structure was June 2028. �
Reuters
This didn't solve the entire balance-sheet problem.
But it changed the immediate question from:
"Could New World face a near-term liquidity crisis?"
to:
"Can management use the additional time to sell assets, generate cash and reduce leverage?"
That's a very important transition.
15. And now we arrive at 2026
This is where the historical comparison becomes fascinating.
Henderson
2026 peak:
HK$35.66
The share price subsequently fell significantly the historical 52-week range reported by ETNet was HK$17.80?HK$35.66 earlier in 2026. �
webuat01.etnet.com.hk
By early August, Henderson was around the high-HK$20s. Morningstar showed HK$26.88 on Aug. 7. �
Morningstar
So Henderson has gone from:
2026 peak ~HK$35.66
→
~HK$26.9
≈ − 25%.
16. New World 2026
New World reached approximately:
HK$12.45
during the 2026 recovery.
But it subsequently fell back toward the:
HK$6?7 range.
Google Finance currently shows a 52-week range of approximately HK$6.09?HK$12.45, with the stock around HK$6?7. �
Google
So:
HK$12.45 → ~HK$6.7
is roughly a:
46% decline
from its 2026 high.
This is a very different risk/reward setup from Henderson.
17. Put the whole history together
Henderson Land
1981 listing
↓
1987 crash
↓
1997 Asian Financial Crisis
↓
2000?03 technology/property downturn
↓
2008 Global Financial Crisis
↓
2009 massive recovery
↓
2011 correction
↓
2015?18 property cycle
↓
2019 Hong Kong crisis
↓
2020 COVID
↓
2021?25 property/rates bear market
↓
2026 recovery
↓
2026 correction
New World
1972 IPO HK$2
↓
1974 global/property downturn
↓
1987 crash
↓
1997 Asian Financial Crisis
↓
2003 SARS
↓
2008 Global Financial Crisis
↓
2009 recovery
↓
2015?18 China/HK property boom
↓
2019 Hong Kong crisis
↓
2020 COVID
↓
2021?25 China/HK property collapse
↓
2024 liquidity/balance-sheet crisis
↓
2025 refinancing
↓
2026 recovery
↓
2026 correction
18. The most important historical comparison
Look at these two episodes:
Henderson 2008?09
− 59.5%
↓
+98.8%
Current cycle
2021?25:
multi-year derating
↓
2026:
initial recovery
↓
2026:
correction
That doesn't guarantee another 2009-style rebound.
But it tells us something important:
The biggest returns in property stocks historically occur when investors buy during the transition from "survival" to "normalization."
Not when the property market is already obviously booming.
19. Henderson versus New World today
This is the most important conclusion from tracing their histories.

Henderson Land
New World
Listing
1981
1972
2026 high
~HK$35.66
~HK$12.45
Current Aug 2026
~HK$27
~HK$6?7
Fall from 2026 high
~25%
~46%
Balance sheet
Much stronger
Much weaker
NAV story
Strong
Potentially enormous
Liquidity risk
Lower
Much higher
Recovery upside
High
Very high
Probability of permanent impairment
Lower
Higher
Investment character
Value
Distressed value
20. Why this matters for your rotation thesis
This historical study actually strengthens the argument we were discussing earlier.
The 2020?25 property sell-off is not unprecedented.
Hong Kong property equities have repeatedly gone through:
boom → leverage → crisis → huge derating → stabilization → NAV recovery → re-rating.
The critical difference this time is that New World has a balance-sheet problem, whereas Henderson is much closer to a conventional property/NAV valuation opportunity.
Therefore I would think of the two stocks differently:
Henderson
2026?27 thesis:
Interest rates stabilize
→ Hong Kong property stabilizes
→ NAV discount narrows
→ developer earnings normalize
→ dividend remains attractive
→ share price re-rates.
New World
2026?28 thesis:
Refinancing buys time
→ asset sales
→ debt reduction
→ property market recovery
→ liquidity risk premium falls
→ NAV discount collapses.
The second scenario has much greater upside if everything works, but also a much greater possibility of permanent capital loss if deleveraging fails.
The historical lesson I would take
The most interesting thing isn't that Henderson and New World have fallen.
It is that they have fallen after an unusually long 2020?25 property/rates bear market, while the market is simultaneously extremely concentrated in AI, technology and banks.
That is precisely the environment in which a long-duration value rotation can eventually begin.
But I would not call the bottom solely from the share price.
For Henderson, I'd watch:
P/NAV + dividend yield + debt cost + Hong Kong property prices.
For New World, I'd put debt/refinancing and asset sales ahead of P/NAV.
And the historical data show why: Henderson's 2008?09 experience demonstrates how violently a quality property stock can rebound after a major derating, while New World's recent history demonstrates that a cheap-looking property stock can remain cheap when leverage becomes the dominant concern. �
Market Cap Of +1
If you want, the next useful step is to � ⁠ build a **1972/1981?2026 crisis-by-crisis table for both companies showing IPO price → crisis low → recovery high → percentage loss → percentage recovery → dividend received**, which would make the comparison much clearer.

Useful To Me Not Useful To Me

chartistkaohz
Supreme

10-Aug-2026 13:57

chartistkaohz About Contact Quote this Post!

x 0
x 0

Alert Admin If by ?how many percent are they buying? you mean how far below the 2026 peak the shares are now, then yes ? we can calculate that. This is separate from the actual percentage ownership bought by European funds I don't see reliable public evidence establishing a specific European-fund purchase percentage for either company.
Using the latest prices available 10 August 2026:
Stock
2026 peak
Current/latest
Fall from 2026 peak
Henderson Land (0012)
HK$35.66
~HK$27.08*
− 24.0%
New World Development (0017)
HK$12.45
~HK$7.23
− 41.9%
*Henderson's HK$35.66 peak occurred 25 Feb 2026 the Financial Times historical data confirms the 2026 high/low dates. � New World's HK$12.45 peak occurred 30 Jan 2026, while the latest intraday quote on Aug. 10 was around HK$7.23. �
Financial Times Markets +1
Financial Times Markets +1
In other words
Henderson Land
HK$35.66 → HK$27.08
= approximately 24% below its 2026 high.
New World
HK$12.45 → HK$7.23
= approximately 42% below its 2026 high.
That difference is significant.
What it means for a European value fund
If a fund is buying after the decline, New World is the much deeper contrarian/distressed-value trade:
Henderson: ~24% below peak
New World: ~42% below peak
But New World carries substantially more balance-sheet/refinancing risk. Reuters reported that Blackstone abandoned its proposed $4 billion transaction in May after disagreements over control, while New World continues working through its debt situation. �
Reuters
Henderson is a much more conventional quality-property/NAV-discount recovery investment. DBS was valuing Henderson at a 47% discount to estimated current NAV in May 2026 and had a HK$35.84 target price. �
DBS Bank
So from the perspective of the European-fund rotation thesis we were discussing:
Henderson = quality value after ~24% correction.
New World = distressed value after ~42% correction.
And that's precisely why a European long-term property fund might be interested: it doesn't need Hong Kong property prices to return to their old highs. A narrowing of the NAV discount alone can generate substantial returns.
.

Useful To Me Not Useful To Me

chartistkaohz
Supreme

10-Aug-2026 13:23

chartistkaohz About Contact Quote this Post!

x 0
x 0

Alert Admin The key idea is not simply ?buy AI.? It is to identify the companies that become unavoidable when AI adoption expands.
For a value-oriented investor, I would divide the AI ecosystem into five layers:
Compute → Power → Physical infrastructure → Security → Industrial automation
The important discovery in 2026 is that the bottleneck is moving downstream from GPUs toward electricity, cooling, grid equipment, cybersecurity and industrial systems.
1. My core thesis
The AI investment chain increasingly looks like this:
AI models ↓
GPUs / custom accelerators ↓
Data centres ↓
Electricity generation ↓
Transmission / transformers / switchgear ↓
Cooling / power management ↓
Cybersecurity ↓
Robotics / automation / smart factories
The further you move down this chain, the less dependent the company is on which particular AI model wins.
That's extremely important.
If Nvidia's next-generation architecture wins, Eaton still sells electrical equipment.
If Google's model wins, Schneider still sells power-management equipment.
If China's AI models become dominant, Siemens can still sell automation equipment.
If OpenAI loses, cybersecurity demand doesn't disappear.
That is what I mean by AI infrastructure with lower technological obsolescence risk.
2. Layer 1 ? AI chips: phenomenal businesses, but valuation matters
The obvious companies are:
NVIDIA
Broadcom
Taiwan Semiconductor Manufacturing Company
Nvidia remains extraordinarily attractive fundamentally. Interestingly, recent market data put its forward P/E around 19x in early August, considerably below some other semiconductor names despite its dominant AI position. �
Barron's
But there's a catch:
AI chips have technological risk.
Today's leader isn't guaranteed to remain the leader.
Competition comes from:
AMD
Google's TPUs
Amazon's custom chips
Microsoft's custom silicon
Chinese accelerators
future Nvidia generations
So the question isn't:
"Will AI chips grow?"
Almost certainly yes.
The question is:
"Who captures the economics of that growth?"
That's why I would rank TSMC and Broadcom differently from speculative AI chip companies.
TSMC benefits from many winners because almost everybody eventually needs advanced semiconductor manufacturing.
3. Layer 2 ? Electricity may be the real bottleneck
This is where the investment thesis becomes much more interesting.
An AI data centre needs:
GPU → server → rack → cooling → power distribution → transformer → grid → electricity generation.
You can't run a 1GW AI cluster simply by ordering more GPUs if the electricity infrastructure isn't available.
Research published this year shows that concentrated AI data-centre construction is already creating regional power-system stress.
And this isn't merely a US issue.
It applies to:
US
Europe
China
Japan
Korea
Singapore
Middle East
That makes electricity infrastructure a global AI investment theme.
4. Eaton: one of my favourite "boring AI" businesses
Eaton is particularly interesting.
Why?
Because Eaton isn't trying to predict which AI model wins.
It sells the infrastructure that allows data centres to operate.
Its Electrical Americas business had:
42% organic order acceleration
48% electrical-sector backlog growth
20% Q1 sales growth
25.6% operating margin
and Eaton raised its 2026 organic-growth guidance to 9?11%. �
Eaton
That's the kind of evidence I like.
The investment equation
AI capex increases
→ data centres increase
→ electricity demand increases
→ electrical infrastructure increases
→ Eaton sells more equipment.
But there is an additional advantage:
Eaton doesn't depend entirely on AI.
It also benefits from:
grid modernization
factories
buildings
aerospace
electrification
industrial automation
So AI becomes an additional growth engine, rather than the entire investment thesis.
My assessment
Business quality: 9/10
AI exposure: 8/10
Diversification: 9/10
Valuation attractiveness: 6/10
Dividend/value characteristics: 6/10
Overall: 8.5/10
I would prefer Eaton after a meaningful valuation correction rather than chase it after a large AI-driven re-rating.
5. GE Vernova: fantastic AI exposure ? but much more valuation risk
GE Vernova is arguably the purest AI → electricity investment.
Its power business benefits when data centres need additional electricity generation.
Its electrification business benefits when that electricity needs to reach the data centre.
2026 revenue guidance was raised to approximately $44.5?45.5 billion, with AI/data-centre demand specifically contributing to power and grid-equipment demand. �
Investing.com
But here's the problem.
The market has already discovered the story.
GE Vernova's market capitalization rose from roughly $39 billion at its 2024 spin-off to $288 billion by July 2026. �
Fortune
Morningstar's July assessment also described the valuation as fair but with very high uncertainty. �
Morningstar
So:
Great company ≠ automatically great stock.
My score
Business quality: 9/10
AI exposure: 9/10
Valuation: 4/10
Dividend/value: 4/10
Risk/reward today: 6/10
I would much rather buy GEV after an AI/power-sector correction than chase it after another parabolic move.
6. Vertiv: probably the highest-beta version of the infrastructure thesis
Vertiv is even more directly connected to AI data centres.
It supplies:
cooling
power systems
thermal management
data-centre infrastructure.
That's excellent.
But the problem is expectations.
Its Q2 results illustrated the issue: earnings were strong and full-year guidance was raised, yet the stock fell sharply because revenue was below what investors expected. �
Reddit
That's classic high-expectation stock behaviour.
It means the market isn't asking:
"Is Vertiv growing?"
It's asking:
"Is Vertiv growing fast enough to justify the valuation?"
That's a much harder hurdle.
My score
Business quality: 9/10
AI exposure: 10/10
Valuation: 5/10
Volatility: 9/10
For a value investor:
Watchlist > chase.
7. Schneider Electric: potentially the best diversified version
Schneider Electric is particularly interesting because it sits between:
AI infrastructure + electricity + industrial automation.
That's a powerful combination.
It provides:
electrical distribution
power management
automation
data-centre infrastructure
industrial controls.
The advantage is diversification.
If AI data-centre construction slows, Schneider still has industrial automation and electrification.
The disadvantage is that investors have already recognized the quality of the business. ABB, for example, was described by Citi earlier this year as fully valued, illustrating how expensive the broader electrification/automation trade has become. �
Investing.com UK
8. Siemens is particularly interesting now
This week's Siemens results are extremely important for your thesis.
Siemens reported its highest-ever quarterly industrial profit.
Industrial profit:
?3.52 billion
↑ 25% YoY
Revenue:
?20.79 billion
↑ 7%
Orders:
?27.90 billion
↑ 13%
And Siemens raised FY2026 EPS guidance to ?11.20??11.50. �
Reuters
But the most interesting part isn't the headline profit.
It's where the demand is coming from.
Siemens says AI-related infrastructure demand is coming from:
data centres
smart factories
electronics manufacturers
industrial AI.
It also works with 9 of the 10 largest data-centre providers, and data-centre orders have increased by triple digits in the first nine months of FY2026. �
Reuters
This is exactly the investment theme I would want:
AI + physical infrastructure + industrial automation.
9. Industrial automation could be the hidden second wave
This is where I think investors may underestimate AI.
AI doesn't merely create data centres.
Eventually it changes factories.
Imagine:
AI → machine vision
→ predictive maintenance
→ robotics
→ automated warehouses
→ autonomous production
→ digital twins
→ smart factories.
That creates demand for:
Siemens
Schneider
ABB
Rockwell Automation
Fanuc
Keyence
Honeywell
The important difference:
AI software
Revenue can potentially be disrupted by a better model.
Industrial automation
The factory still needs:
motors
drives
sensors
controllers
robots
electrical equipment
safety systems.
So AI can actually increase the value of physical automation.
10. This is where US-China competition becomes important
The AI Cold War makes industrial automation even more strategic.
The US wants:
AI leadership + manufacturing reshoring + defence production.
China wants:
AI leadership + manufacturing automation + technological self-sufficiency.
Both therefore need:
robots + automation + smart factories.
This creates an unusual situation where both sides can spend heavily on the same underlying technologies.
That makes companies selling industrial automation potentially more resilient than companies selling one particular AI model.
11. Cybersecurity is the "defence industry" of the AI era
This is perhaps the most important second-order AI trade.
AI agents dramatically increase the number of things that need to be secured.
You don't just have:
employees + computers
You increasingly have:
employees + computers + cloud + APIs + AI agents + autonomous software + industrial systems.
Research this year has already demonstrated substantial improvement in AI agents' ability to perform multi-step cyberattack tasks. �
arXiv
Therefore:
more AI → larger attack surface → greater cybersecurity spending.
That's why I like cybersecurity structurally.
12. Palo Alto Networks is particularly interesting fundamentally
Palo Alto Networks is evolving from a firewall company toward a broad security platform.
Its FY2026 guidance called for:
Revenue: $11.415?11.425B
↑ 24%
Next-generation security ARR:
$8.90?8.95B
↑ 59?60%
Remaining performance obligations:
$20.9?21.0B
↑ 32?33%. �
Palo Alto Networks
Those are extremely strong numbers.
And AI security is becoming an increasingly important product category.
But again:
the stock price matters.
PANW has already recovered dramatically from its 2026 lows one August market report noted a roughly 100% rebound from February lows. �
Investor's Business Daily
So I wouldn't automatically chase it.
13. Fortinet may fit the value philosophy better
Fortinet is interesting because it combines:
hardware + software + cybersecurity + AI infrastructure.
It has also performed strongly this year.
The stock reached $170.35 in July and was still near that level in early August. �
MarketWatch
Again, however, the market knows the story.
The attractive setup would be:
excellent business + temporary earnings disappointment + significant valuation compression
rather than:
excellent business + record valuation + FOMO.
14. Why I prefer cybersecurity over generic AI software
Consider the economics.
AI software
Potential problem:
AI itself may commoditize software.
Cybersecurity
AI creates more things that need protection.
That's almost the opposite economic relationship.
AI is simultaneously:
the threat + the defence.
That makes cybersecurity one of the few areas where increased AI adoption can actually increase the addressable market even if AI model prices fall.
15. The real portfolio architecture
If I were constructing a value-oriented AI basket, I would not put everything into Nvidia.
I'd build it around the infrastructure chain:
Layer
Example
Role
Compute
Nvidia / TSMC
AI engine
Power
Eaton
Electricity infrastructure
Generation
GE Vernova
Power supply
Cooling
Vertiv
Thermal infrastructure
Electrification
Schneider
Grid + data centres
Automation
Siemens
Smart factories
Cybersecurity
Palo Alto
AI defence
Cybersecurity/value
Fortinet
Network security
China AI/automation
Chinese industrial leaders
China ecosystem
16. But there's an important valuation trap
This is the biggest lesson from 2026.
The market has discovered:
"AI needs electricity."
Therefore Eaton, GE Vernova, Vertiv, Schneider and other power stocks have themselves become AI stocks.
That's why buying the "boring" companies isn't automatically value investing anymore.
The market can turn:
boring industrial → AI beneficiary → AI multiple → expensive stock.
GE Vernova is the clearest example.
So I would use a two-stage filter.
Stage 1: Strategic importance
Does AI fundamentally increase demand?
Stage 2: Valuation
Has the market already capitalized that demand?
Only buy when both are favourable.
17. My preferred hierarchy today
If I rank the business models, rather than today's share prices:
🥇 Eaton
Best combination of AI + electricity + diversification
🥈 Siemens
Excellent AI + automation + industrial exposure
🥉 TSMC
Broadest semiconductor exposure across competing AI ecosystems
4. Palo Alto Networks
Excellent AI/cybersecurity structural growth
5. Schneider Electric
Excellent electrification + automation + data centres
6. Fortinet
Cybersecurity + infrastructure with attractive business economics
7. Nvidia
Best AI semiconductor franchise, but valuation/expectations matter
8. Vertiv
Excellent AI infrastructure exposure, but higher valuation sensitivity
9. GE Vernova
Excellent AI-power beneficiary, but valuation is the major issue
18. The most interesting part for a Singapore investor
There is another layer.
Singapore doesn't need to produce the world's best AI model to benefit.
Singapore can benefit from:
AI → data centres
→ electricity
→ cooling
→ cybersecurity
→ financial infrastructure
→ semiconductor ecosystem
→ regional headquarters.
That means the broader Singapore/Asian infrastructure ecosystem can benefit even if the US-China AI competition remains unresolved.
And Hong Kong/China can participate through a different channel:
Chinese AI → industrial automation
→ EVs
→ robotics
→ manufacturing
→ semiconductor localization.
So I would not view this as simply:
US AI vs China AI
I would view it as:
US AI ecosystem + China AI ecosystem + the infrastructure providers supplying both.
19. My biggest investment conclusion
The first AI wave was:
Buy the model / chip winner.
The second AI wave is:
Buy the infrastructure required by all winners.
And the third AI wave could be:
Buy the companies that use AI to transform the physical economy.
That gives us:
Wave 1
Nvidia / Broadcom / TSMC
Wave 2
Eaton / Schneider / Vertiv / GE Vernova / cybersecurity
Wave 3
Siemens / ABB / Fanuc / industrial automation / robotics
For a value-oriented investor, Wave 2 and Wave 3 are particularly interesting, because they can potentially capture AI growth without having to predict which LLM wins.
My preferred strategy
I would not chase this week's AI rally.
I'd build a watchlist and wait for valuation-driven entry points.
The ideal scenario is:
AI spending remains strong + company earnings continue rising + stock falls 20?30% because expectations temporarily disappoint
much better risk/reward.
That's exactly the type of situation where a long-term, valuation-conscious investor has an advantage over momentum investors.
The current Siemens results are a good illustration: fundamentals are accelerating even while the share price fell 5.2% after the results, showing the type of disconnect worth monitoring. �
Reuters
In other words, I would rather own a great AI infrastructure company at 20×?25× normalized earnings than a fashionable AI software company at 50×?80× earnings. The AI arms race can continue for a decade that doesn't mean we have to pay today's maximum price for it.

Useful To Me Not Useful To Me

chartistkaohz
Supreme

10-Aug-2026 13:17

chartistkaohz About Contact Quote this Post!

x 0
x 0

Alert Admin ? the comparison to the US?Soviet Cold War is useful, but there is one major difference: the AI rivalry is much more commercially integrated. The US and China are competing while still buying, selling, investing and trading with the rest of the world.
The current evidence suggests this is becoming a long-duration technological competition, not simply a temporary trade dispute. China is challenging the US in both AI hardware and software, while Washington is trying to preserve its lead through semiconductor controls and a broader strategy to export a US-centered AI stack to allies. �
Bruegel +1
1. The 1980s Cold War analogy
Think of the competition as:
1980s Cold War
2026 AI competition
US vs USSR
US vs China
Nuclear weapons
Advanced AI + compute
ICBMs
AI chips/data centres
Space race
AI capability race
Military electronics
AI + robotics + autonomous systems
Satellites
Cloud/AI infrastructure
NATO/allies
US-aligned AI ecosystem
Soviet bloc
China-centered technology ecosystem
Arms race
AI investment race
Export controls
Chip/export controls
Intelligence/cyber
AI + cybersecurity
But there is an important difference.
The Soviet Union was economically much less integrated with the US-led global economy. China isn't.
China is deeply integrated into global manufacturing, supply chains, commodities, electronics and capital markets.
That makes today's AI competition potentially more economically powerful but also more complicated.
2. The biggest investment consequence: the world may build TWO AI ecosystems
This is probably the most important long-term investment implication.
US ecosystem
US:
Nvidia → TSMC → Microsoft → OpenAI → Google → Amazon → Meta → cloud → cybersecurity
Washington is also actively promoting exports of integrated US AI technology packages to international partners, including AI hardware, data systems and AI models. �
Trade.gov +1
China ecosystem
China:
Huawei → Chinese semiconductor ecosystem → Alibaba/Tencent/Baidu/ByteDance → Chinese AI models → Chinese cloud → domestic applications
China is increasingly emphasizing domestic AI capability and open AI ecosystems as US restrictions make access to advanced computing more difficult. Recent research argues that US restrictions may actually have accelerated China's development of locally adaptable/open AI systems. �
arXiv
So instead of:
One global AI market
we could eventually get:
US-led AI + China-led AI + a large neutral/global middle
That is extremely important for investors.
3. The AI arms race could actually increase global AI spending
This is where I differ from the simple "AI bubble" argument.
If the US believes China is catching up, Washington has an incentive to spend more, not less.
If China believes the US is trying to contain it, Beijing has an incentive to spend more, not less.
That creates an unusual investment dynamic:
US government
More spending on:
AI
semiconductors
data centres
electricity
nuclear power
cybersecurity
military AI
robotics
quantum computing
China government
More spending on:
domestic chips
AI models
data centres
electricity
robotics
industrial automation
autonomous vehicles
AI applications
semiconductor equipment
So even if individual AI stocks experience a 30?50% valuation correction, the underlying strategic investment cycle could continue.
4. This is why I would separate the AI investment universe into layers
Tier 1 ? AI "picks and shovels"
Potentially the most durable part of the cycle:
Semiconductors + semiconductor equipment + networking + power + data centres
Examples include:
Nvidia
AMD
Broadcom
TSMC
ASML
Micron
SK Hynix
Applied Materials
The problem is valuation.
A fantastic business can still be a bad investment if you pay too much.
5. Tier 2 ? Cloud and hyperscalers
This is potentially more attractive from a risk/reward perspective.
Think:
Microsoft / Amazon / Alphabet / Meta
They have enormous existing cash flows that can finance AI spending.
The strategic advantage is that they don't need AI alone to justify their valuations.
For example:
Microsoft can spend billions on AI while still having Windows, Office, Azure and enterprise software.
That's very different from a small AI company whose entire valuation depends on one AI product.
6. Tier 3 ? AI cybersecurity
This is the area I find particularly interesting from the Cold War analogy.
The more AI capabilities increase, the more valuable cybersecurity becomes.
Why?
AI can be used by both attackers and defenders.
So you get:
AI → more sophisticated attacks → greater security spending → more AI cybersecurity
That creates a potential positive feedback loop.
The Cold War analogy would be:
Nuclear weapons increased the importance of missile defence.
Similarly:
AI increases the importance of AI security.
This means cybersecurity could become a structural beneficiary of the AI arms race, rather than merely another AI software trade.
7. Tier 4 ? AI software is much more complicated
This is where I would be careful.
AI could simultaneously:
Help software companies
AI makes software more powerful.
But AI could also:
Destroy software pricing
If an AI agent can perform tasks that previously required five different SaaS applications, some software companies could lose pricing power.
That's why the recent AI software sell-off is actually economically important.
The market is asking:
Will AI increase software revenue, or replace software?
The answer will probably be different for every company.
8. China creates another huge investment opportunity
This is where your interest in Hong Kong stocks becomes particularly relevant.
China's AI development doesn't necessarily require China to beat Nvidia immediately.
China can make money from AI through:
industrial automation
robotics
EVs
autonomous driving
logistics
e-commerce
financial technology
smartphones
manufacturing
healthcare
smart factories
China has enormous industrial scale.
Therefore:
China may monetize AI through physical industry rather than purely through AI software.
That could become a major investment theme.
9. The biggest winner might actually be electricity
This is an underappreciated consequence.
AI requires enormous amounts of:
chips → data centres → electricity → cooling → transmission infrastructure
Therefore the AI arms race becomes partly an energy arms race.
The US needs:
gas
nuclear
renewables
transmission
transformers
grid infrastructure
China has a major advantage in electricity generation and industrial infrastructure.
That means AI investment eventually spreads far beyond technology.
10. And then comes cybersecurity
Imagine the geopolitical structure:
US AI ↓
Chinese AI competition
↓
More cyber competition
↓
More attacks/defence requirements
↓
More cybersecurity spending
↓
More AI used in cybersecurity
↓
More AI compute required
That creates a powerful investment ecosystem.
So I would not think of:
AI + cybersecurity
as two completely separate themes.
They are increasingly interconnected.
11. What happens to global markets?
I see roughly three scenarios.
Scenario A ? Controlled competition ⭐ ⭐ ⭐ ⭐ ⭐
US and China compete intensely but avoid a complete technological divorce.
Result:
AI capex remains enormous
US companies maintain leadership
Chinese AI improves rapidly
global AI adoption continues
Europe/Japan/Korea/Singapore/Taiwan remain important
Best investment environment.
Scenario B ? Technology bifurcation ⭐ ⭐ ⭐ ⭐
US and China create largely separate technology ecosystems.
Result:
Higher duplication of infrastructure.
The world might effectively build:
US AI infrastructure
AND
China AI infrastructure
That could actually mean more total AI capital expenditure.
Companies supplying both ecosystems could benefit, but geopolitical risk becomes much higher.
Scenario C ? Severe technology war ⭐ ⭐
Export controls escalate dramatically.
China accelerates domestic semiconductor development.
US restricts Chinese access to increasingly broad AI technology.
China retaliates through critical minerals, manufacturing and other supply chains.
This would probably create:
higher semiconductor prices
supply-chain disruption
lower corporate margins
market volatility
lower valuations
But ironically, it could also increase government spending on domestic technology.
12. The biggest difference from the 1980s
There is one huge advantage for investors today.
During the Cold War, technology was concentrated heavily inside governments and military contractors.
Today:
AI is simultaneously a military technology, commercial technology and consumer technology.
That means the economic opportunity is much larger.
AI isn't just about weapons.
It's:
AI → cloud → software → cybersecurity → chips → robotics → factories → energy → financial services → healthcare → transportation.
That's why I think the AI investment cycle could be much longer than the market currently assumes.
13. But don't confuse "AI arms race" with "every AI stock goes up"
This is the critical investment lesson.
The Cold War created enormous technological advances, but not every defence company became a great investment.
Likewise:
AI can be enormously successful while many AI stocks are terrible investments.
The winners will probably be companies with:
high ROE + recurring cash flow + pricing power + reasonable valuation + strategic importance.
That fits much better with your value/dividend approach than simply chasing the highest-growth AI names.
My long-term AI investment hierarchy
If I were constructing the theme from a value-investor perspective, I'd think about it like this:
Theme
Strategic importance
Valuation risk
AI chips
⭐ ⭐ ⭐ ⭐ ⭐
🔴 Very high
Semiconductor equipment
⭐ ⭐ ⭐ ⭐ ⭐
🔴 High
Cloud
⭐ ⭐ ⭐ ⭐ ⭐
🟠 Medium-high
Cybersecurity
⭐ ⭐ ⭐ ⭐ ⭐
🟠 Medium-high
AI software
⭐ ⭐ ⭐ ⭐
🔴 Very high
Data centres
⭐ ⭐ ⭐ ⭐ ⭐
🟠 Medium
Electricity/grid
⭐ ⭐ ⭐ ⭐ ⭐
🟢 Lower
Robotics
⭐ ⭐ ⭐ ⭐ ⭐
🟠 Medium-high
Chinese AI/automation
⭐ ⭐ ⭐ ⭐ ⭐
🟠 /🟢 Depends heavily on valuation
The big conclusion
I would not treat the current AI sell-off simply as the end of the AI boom.
I would treat it more like a Cold War-style transition from an initial technology boom toward a much larger strategic infrastructure build-out.
The key question is changing from:
"Which company has the best AI model?"
to:
"Who controls the compute, chips, electricity, cloud, cybersecurity and industrial applications that make AI economically useful?"
That second question is much more important for a long-term investor.
And China's rapid progress is particularly important: recent analysis suggests Chinese AI is increasingly competing not only through hardware but also through lower-cost models and software ecosystems. �
thinkchina.sg +1
**For your portfolio style, I would therefore be much more interested in finding the "AI infrastructure + cybersecurity + electricity + industrial automation" winners at reasonable valuations than simply buying the most fashionable AI software stocks.**

Useful To Me Not Useful To Me

chartiskao
Supreme

05-Aug-2026 16:27

chartiskao About Contact Quote this Post!

x 0
x 0

Alert Admin

Tencent (0700.HK) is one of the world' s most profitable AI platform companies, although whether it is the most profitable depends on the metric used. Here' s a balanced assessment.

Executive Verdict

Category	Rating
Business Quality	⭐ ⭐ ⭐ ⭐ ⭐ (10/10)
Profitability	⭐ ⭐ ⭐ ⭐ ⭐ (10/10)
AI Monetization	⭐ ⭐ ⭐ ⭐ ⭐ (9.8/10)
Financial Strength	⭐ ⭐ ⭐ ⭐ ⭐ (9.7/10)
Valuation	⭐ ⭐ ⭐ ⭐ ⭐ (9.5/10)
Long-term Outlook	⭐ ⭐ ⭐ ⭐ ⭐ (Excellent)

Overall, Tencent combines strong cash generation, high margins, and a diversified AI ecosystem in a way that very few companies globally can match.

1. AI Companies: Revenue vs. Profit

Many leading AI companies are still investing heavily and accept lower near-term profitability. Tencent stands out because its AI investments are supported by already profitable businesses.

Company	Approximate Situation
Microsoft	Extremely profitable AI integrated across products
Meta Platforms	Highly profitable AI driving advertising
Alphabet	Highly profitable AI across Search and Cloud
Amazon	Strong profits led by AWS retail margins lower
Tencent Holdings	Highly profitable AI integrated into gaming, WeChat, advertising, fintech and cloud
OpenAI	Rapid revenue growth but still investing heavily

Tencent belongs in the same tier as Microsoft, Alphabet, and Meta rather than AI companies that are still building sustainable profitability.

2. Profit Margin Above 30%

A net profit margin of roughly 30.6% is exceptional.
For every RMB100 of revenue:

Around RMB30 becomes net profit.
Few companies at Tencent' s scale consistently achieve this level of profitability.

This reflects the strength of its software, gaming, advertising, and fintech businesses.

3. Operating Margin

Operating margin:
34.3%
This indicates Tencent generates strong profits before financing costs and taxes while continuing to invest heavily in AI.

4. Gross Margin

Gross margin:
56.5%
This means Tencent keeps more than half of every yuan of revenue after direct operating costs.
Businesses with gross margins above 50% often benefit from scalable software and platform economics.

5. Return on Equity (ROE)

ROE:
20.5%
Many investors consider:

Above 15% = strong.
Above 20% = excellent.

Tencent is generating attractive returns on shareholders' capital.

6. Cash Position

Total cash:
RMB459.7 billion
Debt:
RMB405.5 billion
Although Tencent carries debt, it also has a substantial cash balance and strong operating cash flow, giving it flexibility to invest, repurchase shares, or pursue acquisitions.

7. Cash Generation

Operating cash flow:
RMB327.5 billion
This is one of Tencent' s greatest strengths.
It means the company can fund:

AI chips
Data centers
AI talent
New games
Cloud expansion
Share buybacks
Dividends

without depending heavily on external financing.

8. Valuation

Current valuation metrics are notably lower than in previous years.

Metric	Current	Historical Range (shown)
Trailing P/E	17.2	22&ndash 27
Forward P/E	13.9	17&ndash 20
P/S	5.1	6.6&ndash 8.0
P/B	3.3	4.0&ndash 4.9

If earnings continue to grow while valuation multiples remain compressed, shareholders can benefit from both earnings growth and potential multiple expansion.

9. AI Is Enhancing Existing Businesses

Unlike companies that must build entirely new revenue streams, Tencent is embedding AI into products that already have large user bases.
Examples include:

WeChat recommendations and services.
Advertising optimization.
Game development.
Customer support.
Enterprise cloud offerings.

This approach may provide steadier monetization than relying on standalone AI products alone.

10. Balance Sheet

Key indicators suggest a healthy financial position:

Current ratio above 1.
Debt-to-equity around one-third.
Strong operating cash flow.
Significant cash reserves.

That combination provides resilience during periods of elevated AI investment.

Risks

Despite these strengths, investors should continue to watch:

Competition in AI from Chinese and global technology companies.
Regulatory developments in China.
U.S. export controls affecting advanced AI hardware.
The pace at which AI investments translate into incremental earnings.

These factors could influence future growth and valuation.

Is Tencent undervalued?

Compared with its own recent history, the valuation metrics you listed are lower even as profitability remains strong. That can indicate a more attractive valuation than in prior years, though it may also reflect investor concerns about China' s economy, regulation, or geopolitical risks.
Whether it is " undervalued" ultimately depends on future earnings growth, but the combination of:

forward P/E around 12&ndash 14,
ROE above 20%,
net margins above 30%,
robust cash generation,
and a diversified AI ecosystem

is uncommon among global technology leaders.

Overall assessment

Tencent' s investment case is built on more than AI alone. Its gaming, advertising, fintech, cloud, and WeChat ecosystem generate substantial cash today, while AI is being integrated across those businesses to improve products and efficiency. That gives Tencent an advantage over companies that must first prove they can monetize AI.
Calling Tencent the world' s most profitable AI company would overstate the evidence because companies such as Microsoft and Alphabet generate higher absolute profits while also being major AI leaders. A more supportable conclusion is that Tencent is one of the world' s most profitable AI platform companies and is currently trading at valuation multiples that are lower than many global AI peers despite maintaining high profitability and significant investment capacity.

chartiskao ( Date: 05-Aug-2026 16:21) Posted:

A summary highlights an important but sometimes overlooked aspect of the AI competition: the battle is not only for GPUs and data centers, but also for elite AI researchers and engineers. Tencent' s numbers suggest management believes that human talent is one of its most valuable long-term investments.

Key Takeaways

1. Nearly 60% increase in share-based compensation

Tencent' s share-based compensation rose to approximately RMB 7.46 billion in the first half of 2025, up nearly 60% year over year.
This tells investors that Tencent is deliberately using stock-based incentives to:

Recruit top AI researchers.
Retain existing engineers.
Prevent competitors from poaching key staff.
Align employees' interests with long-term shareholder value.

Rather than simply increasing cash salaries, equity awards encourage employees to stay and contribute over multiple years.

2. Human capital is Tencent' s largest R& D investment

You noted that around 77% of R& D spending goes toward:

Salaries
Employee benefits
Bonuses
Research staff

This is typical of leading AI companies.
Unlike traditional manufacturing, AI breakthroughs come primarily from highly skilled people rather than physical assets.
The formula is roughly:

Better researchers &rarr Better models &rarr Better products &rarr More users &rarr Higher revenue

3. AI talent is extremely scarce

Tencent is competing globally for AI experts with companies such as:

OpenAI
Google DeepMind
Anthropic
Meta Platforms
Microsoft
Alibaba Group
ByteDance
Baidu

Experienced AI researchers can command compensation packages worth millions of dollars annually because demand far exceeds supply.

4. Long-term incentive plans

Tencent' s 2023 Share Option Scheme and Share Award Scheme are designed to reward employees over time.
These programs generally:

Grant shares or options that vest over several years.
Encourage employees to remain with the company.
Tie rewards to the company' s long-term performance.

This helps reduce turnover in a highly competitive labor market.

Why this matters for shareholders

Although higher share-based compensation is an expense that can dilute existing shareholders if new shares are issued, it can be worthwhile if it helps Tencent:

Build better AI models.
Launch successful AI-powered products.
Improve advertising performance.
Increase cloud and enterprise AI revenue.
Enhance gaming development and user engagement.

If these investments generate returns that exceed their cost, shareholders ultimately benefit through stronger earnings growth.

Comparing Tencent with other AI leaders

Many of the world' s leading AI companies are making similar investments:

Company	AI Talent Strategy
Meta Platforms	Aggressive hiring, large stock awards, major AI recruiting
Microsoft	Significant equity compensation and strategic AI partnerships
Alphabet	Heavy investment in researchers through Google DeepMind
Amazon	Expanding AI teams across cloud and consumer businesses
Tencent	Rapidly increasing equity incentives and AI-focused R& D hiring

Tencent' s approach is therefore consistent with the broader industry trend among companies aiming to lead in generative AI.

Investor interpretation

From an investor' s perspective, these figures suggest Tencent is pursuing a balanced strategy:

Infrastructure: investing heavily in GPUs and data centers.
People: investing heavily in elite AI researchers and engineers.
Products: integrating its Hunyuan models into WeChat, gaming, advertising, fintech, and enterprise services to generate commercial returns.

The combination of capital investment, talent acquisition, and deployment into existing high-margin businesses gives Tencent a stronger foundation than companies that are investing heavily in AI but have not yet demonstrated meaningful monetization. If Tencent continues to convert these investments into higher revenue and profits while maintaining healthy margins, the increased spending on AI talent could prove to be a significant long-term competitive advantage.

chartiskao ( Date: 05-Aug-2026 15:59) Posted:

https://www.youtube.com/watch?v=2o5suWtdTRM

Executive Summary

Overall Rating: Very Strong (9.5/10)
Tencent is demonstrating three things simultaneously:

Strong double-digit revenue growth.
Expanding profit margins despite heavy AI investment.
Ability to fund AI expansion entirely from operating cash flow.

This combination is uncommon. Many AI companies are sacrificing profits to invest, whereas Tencent is growing both earnings and investment capacity.

1. Revenue Growth (15%)

Revenue reached RMB184.5 billion, up 15% YoY.
That is impressive because Tencent is already one of China' s largest companies.
For comparison:

5% growth is respectable for a mega-cap.
10% is considered strong.
15% is exceptional at Tencent' s scale.

The growth came from multiple businesses rather than one-off gains.

Business	Growth
Gaming	Strong
Advertising	Strong
FinTech	Stable
Cloud	Improving
AI	Beginning monetization

This makes Tencent less dependent on any single business.

2. Operating Leverage Is Improving

Gross profit:
+22%
Revenue:
+15%
Gross margin:
53% &rarr 57%
This means costs are growing more slowly than revenue.
For every additional RMB100 earned, Tencent is keeping more profit than before.
That is one of the best signs investors can see.

3. Core Business Growing 20%

Management highlighted:

Excluding investment gains and losses, the operating businesses grew around 20%.

That means:

WeChat
Advertising
FinTech
Games

are becoming even stronger.
This is healthier than relying on investment portfolio gains.

4. Gaming Remains a Cash Machine

Gaming still contributes roughly half of revenue.

Domestic Games

Games like

Honor of Kings

continue generating enormous cash flow.
These " evergreen" titles are similar to recurring subscription businesses.

International Gaming

Revenue rose:
35%
This is especially encouraging.
Tencent owns or controls stakes in many international studios.
Growth overseas reduces dependence on China' s domestic market.

5. WeChat Is Still Tencent' s Biggest Competitive Advantage

WeChat now has:
1.4 billion monthly active users
Very few companies possess such a platform.
People use WeChat for:

Messaging
Payments
Shopping
Mini Programs
Government services
Banking
Transport
Entertainment

This creates an ecosystem competitors struggle to replicate.

6. AI Is Already Generating Revenue

Many companies are spending heavily on AI without clear returns.
Tencent appears further along.
Examples include:

Advertising

Revenue:
+20%
Management credits AI for:

Better ad targeting
Better recommendation engines
Improved creative generation

This directly increases advertiser ROI.

Customer Service

AI handles merchant inquiries.
Benefits:

Lower labour costs
Faster responses
Better customer satisfaction

Gaming

AI helps developers:

Design characters
Generate dialogue
Test software
Create content

This reduces production costs and shortens development cycles.

7. Massive AI Investment

Capex:
RMB19.1 billion
Up:
120%
That' s enormous.
The spending includes:

NVIDIA GPUs
AI servers
Data centres
Networking equipment

Tencent is investing before AI demand peaks.

8. Hiring AI Talent

Share-based compensation increased nearly:
60%
High-end AI researchers are extremely scarce.
Tencent is competing with:

Alibaba Group
ByteDance
Baidu
DeepSeek

for elite engineers.
This investment could strengthen Tencent' s competitive position over the coming years.

9. Why Margins Matter

Usually companies investing heavily in AI see margins fall.
Tencent achieved:
Higher investment
AND
Higher margins.
That suggests:

Excellent cost control.
Strong pricing power.
High-quality businesses funding AI expansion.

10. Comparison with Global AI Leaders

Company	AI Spending	Current Profitability
Microsoft	Very High	Very High
Meta Platforms	Very High	Very High
Amazon	Very High	Moderate
Tencent	Very High	Very High

Tencent resembles Microsoft and Meta more than companies still searching for profitable AI business models.

Risks

Investors should still monitor:

Slower Chinese consumer spending.
Regulatory changes affecting internet platforms.
U.S. export controls on advanced AI chips.
Competition in AI models and applications.
Gaming approval timelines.

These factors could affect growth or margins.

Long-Term Outlook (2026&ndash 2030)

Tencent appears to be entering a new growth phase where its existing businesses generate substantial cash while AI enhances products and opens new revenue opportunities.
If management continues executing well, the combination of:

WeChat' s massive ecosystem,
leading gaming franchises,
growing advertising business,
fintech platform,
and sustained AI investment

could support faster earnings growth over the next several years.
Based on the figures you' ve provided, these results suggest Tencent is not merely investing in AI for future potential&mdash it is already seeing measurable commercial benefits while continuing to strengthen its financial performance. For long-term investors, that is one of the most encouraging aspects of this earnings report.

Useful To Me Not Useful To Me

chartiskao
Supreme

05-Aug-2026 16:21

chartiskao About Contact Quote this Post!

x 0
x 0

Alert Admin

Key Takeaways

1. Nearly 60% increase in share-based compensation

Recruit top AI researchers.
Retain existing engineers.
Prevent competitors from poaching key staff.
Align employees' interests with long-term shareholder value.

Rather than simply increasing cash salaries, equity awards encourage employees to stay and contribute over multiple years.

2. Human capital is Tencent' s largest R& D investment

You noted that around 77% of R& D spending goes toward:

Salaries
Employee benefits
Bonuses
Research staff

This is typical of leading AI companies.
Unlike traditional manufacturing, AI breakthroughs come primarily from highly skilled people rather than physical assets.
The formula is roughly:

Better researchers &rarr Better models &rarr Better products &rarr More users &rarr Higher revenue

3. AI talent is extremely scarce

Tencent is competing globally for AI experts with companies such as:

OpenAI
Google DeepMind
Anthropic
Meta Platforms
Microsoft
Alibaba Group
ByteDance
Baidu

Experienced AI researchers can command compensation packages worth millions of dollars annually because demand far exceeds supply.

4. Long-term incentive plans

Tencent' s 2023 Share Option Scheme and Share Award Scheme are designed to reward employees over time.
These programs generally:

Grant shares or options that vest over several years.
Encourage employees to remain with the company.
Tie rewards to the company' s long-term performance.

This helps reduce turnover in a highly competitive labor market.

Why this matters for shareholders

Although higher share-based compensation is an expense that can dilute existing shareholders if new shares are issued, it can be worthwhile if it helps Tencent:

Build better AI models.
Launch successful AI-powered products.
Improve advertising performance.
Increase cloud and enterprise AI revenue.
Enhance gaming development and user engagement.

If these investments generate returns that exceed their cost, shareholders ultimately benefit through stronger earnings growth.

Comparing Tencent with other AI leaders

Many of the world' s leading AI companies are making similar investments:

Company	AI Talent Strategy
Meta Platforms	Aggressive hiring, large stock awards, major AI recruiting
Microsoft	Significant equity compensation and strategic AI partnerships
Alphabet	Heavy investment in researchers through Google DeepMind
Amazon	Expanding AI teams across cloud and consumer businesses
Tencent	Rapidly increasing equity incentives and AI-focused R& D hiring

Tencent' s approach is therefore consistent with the broader industry trend among companies aiming to lead in generative AI.

Investor interpretation

From an investor' s perspective, these figures suggest Tencent is pursuing a balanced strategy:

Infrastructure: investing heavily in GPUs and data centers.
People: investing heavily in elite AI researchers and engineers.
Products: integrating its Hunyuan models into WeChat, gaming, advertising, fintech, and enterprise services to generate commercial returns.

chartiskao ( Date: 05-Aug-2026 15:59) Posted:

https://www.youtube.com/watch?v=2o5suWtdTRM

Executive Summary

Overall Rating: Very Strong (9.5/10)
Tencent is demonstrating three things simultaneously:

Strong double-digit revenue growth.
Expanding profit margins despite heavy AI investment.
Ability to fund AI expansion entirely from operating cash flow.

This combination is uncommon. Many AI companies are sacrificing profits to invest, whereas Tencent is growing both earnings and investment capacity.

1. Revenue Growth (15%)

Revenue reached RMB184.5 billion, up 15% YoY.
That is impressive because Tencent is already one of China' s largest companies.
For comparison:

5% growth is respectable for a mega-cap.
10% is considered strong.
15% is exceptional at Tencent' s scale.

The growth came from multiple businesses rather than one-off gains.

Business	Growth
Gaming	Strong
Advertising	Strong
FinTech	Stable
Cloud	Improving
AI	Beginning monetization

This makes Tencent less dependent on any single business.

2. Operating Leverage Is Improving

3. Core Business Growing 20%

Management highlighted:

Excluding investment gains and losses, the operating businesses grew around 20%.

That means:

WeChat
Advertising
FinTech
Games

are becoming even stronger.
This is healthier than relying on investment portfolio gains.

4. Gaming Remains a Cash Machine

Gaming still contributes roughly half of revenue.

Domestic Games

Games like

Honor of Kings

continue generating enormous cash flow.
These " evergreen" titles are similar to recurring subscription businesses.

International Gaming

Revenue rose:
35%
This is especially encouraging.
Tencent owns or controls stakes in many international studios.
Growth overseas reduces dependence on China' s domestic market.

5. WeChat Is Still Tencent' s Biggest Competitive Advantage

WeChat now has:
1.4 billion monthly active users
Very few companies possess such a platform.
People use WeChat for:

Messaging
Payments
Shopping
Mini Programs
Government services
Banking
Transport
Entertainment

This creates an ecosystem competitors struggle to replicate.

6. AI Is Already Generating Revenue

Many companies are spending heavily on AI without clear returns.
Tencent appears further along.
Examples include:

Advertising

Revenue:
+20%
Management credits AI for:

Better ad targeting
Better recommendation engines
Improved creative generation

This directly increases advertiser ROI.

Customer Service

AI handles merchant inquiries.
Benefits:

Lower labour costs
Faster responses
Better customer satisfaction

Gaming

AI helps developers:

Design characters
Generate dialogue
Test software
Create content

This reduces production costs and shortens development cycles.

7. Massive AI Investment

Capex:
RMB19.1 billion
Up:
120%
That' s enormous.
The spending includes:

NVIDIA GPUs
AI servers
Data centres
Networking equipment

Tencent is investing before AI demand peaks.

8. Hiring AI Talent

Share-based compensation increased nearly:
60%
High-end AI researchers are extremely scarce.
Tencent is competing with:

Alibaba Group
ByteDance
Baidu
DeepSeek

for elite engineers.
This investment could strengthen Tencent' s competitive position over the coming years.

9. Why Margins Matter

Usually companies investing heavily in AI see margins fall.
Tencent achieved:
Higher investment
AND
Higher margins.
That suggests:

Excellent cost control.
Strong pricing power.
High-quality businesses funding AI expansion.

10. Comparison with Global AI Leaders

Company	AI Spending	Current Profitability
Microsoft	Very High	Very High
Meta Platforms	Very High	Very High
Amazon	Very High	Moderate
Tencent	Very High	Very High

Tencent resembles Microsoft and Meta more than companies still searching for profitable AI business models.

Risks

Investors should still monitor:

Slower Chinese consumer spending.
Regulatory changes affecting internet platforms.
U.S. export controls on advanced AI chips.
Competition in AI models and applications.
Gaming approval timelines.

These factors could affect growth or margins.

Long-Term Outlook (2026&ndash 2030)

WeChat' s massive ecosystem,
leading gaming franchises,
growing advertising business,
fintech platform,
and sustained AI investment

chartiskao ( Date: 05-Aug-2026 15:56) Posted:

https://www.youtube.com/watch?v=lFqDC4lOUqc

Thomas (Tom) Lee is one of the best-known U.S. equity strategists on Wall Street. He co-founded Fundstrat Global Advisors after spending about 25 years as a highly ranked equity strategist at JPMorgan Chase. He is well known for being consistently optimistic (" bullish" ) on U.S. stocks, particularly technology, AI, and Bitcoin.

Why investors pay attention to Tom Lee

Lee has built a reputation because:
Successful calls

Bullish after the COVID-19 crash in 2020.
Correctly anticipated the strong U.S. bull market in 2023&ndash 2025 while many strategists remained cautious.
Has consistently argued that AI would drive a multi-year earnings boom.

Mixed record
Like every market strategist, he has also made incorrect calls. Some of his bullish forecasts arrived earlier than the market was ready for. Investors generally view him as someone who is directionally positive over the long term rather than someone who times every correction perfectly.

Why does he think 2027 could be an outstanding year?

Based on the comments you summarized, Lee believes several temporary headwinds should disappear by then.

1. AI spending begins generating profits

Today, markets worry that companies are spending hundreds of billions on AI infrastructure.
Examples include:

Microsoft
Amazon
Meta Platforms
Alphabet

Lee believes:
2025&ndash 2026

Massive capital expenditure.
Investors worry about returns.

2027

AI applications begin producing significant revenue.
Corporate earnings accelerate.
Investors reward companies that invested early.

This is similar to how internet investments in the late 1990s eventually produced large profits years later.

2. Fed uncertainty disappears

Currently markets are debating:

How much the Federal Reserve will cut rates.
Whether inflation stays under control.
Whether economic growth slows.

Lee expects by 2027:

Inflation is largely controlled.
Interest rates become more predictable.
Companies can plan investments with greater confidence.

Lower uncertainty usually supports higher stock valuations.

3. Housing inflation cools

Housing makes up a large part of U.S. inflation.
Lee points to:

Falling home prices in some regions.
Slower rent increases.
Softer wage growth.

If those trends continue:

Overall inflation declines.
The Fed has more flexibility to ease policy.
Stocks often respond positively.

4. Forced selling is temporary

Lee mentioned several market disruptions.
These include:

AI-related profit taking.
Leveraged investors being forced to sell.
Deleveraging by hedge funds.
Volatility linked to South Korea and semiconductor stocks.

When leverage is unwound:

Selling pressure eventually ends.
Quality companies can recover.
New buyers return.

He refers to this as the market being like a " coiled spring."

5. Strong earnings cycle

Lee expects:

AI productivity to improve profits.
Economic growth to remain positive.
S& P 500 earnings to keep rising.

Historically, stock prices follow earnings over long periods.
If earnings grow 10&ndash 15% annually, indexes can continue reaching new highs.

Why did semiconductor stocks fall?

Lee attributes much of the decline to forced liquidation, not deteriorating business fundamentals.
He specifically mentioned:

Samsung Electronics
SK Hynix

The idea is that leveraged investors sold these holdings to reduce risk, creating sharp short-term declines even if long-term demand for AI chips remained intact.

Is Tom Lee' s view reasonable?

There is logic behind his argument:
Bullish factors

AI investment eventually translates into earnings.
Lower inflation can support lower interest rates.
Reduced policy uncertainty can improve investor confidence.
Corporate profits remain the primary long-term driver of stock prices.

However, there are meaningful risks:

AI spending may not generate the expected returns.
A recession could reduce earnings.
Geopolitical tensions (for example, U.S.&ndash China relations) could disrupt technology supply chains.
Inflation could reaccelerate, limiting the Fed' s ability to ease.

What does this mean for your investment approach?

Given your focus on long-term investing, dividends, and value, Lee' s outlook broadly aligns with a patient strategy:

If AI drives a sustained earnings cycle, large technology companies and semiconductor firms could benefit over the next several years.
Lower interest rates would also tend to support banks, REITs, property developers, and insurers&mdash sectors you' ve been following.
At the same time, maintaining diversification remains important because Lee' s 2027 scenario depends on several macroeconomic assumptions that may not all materialize.

In other words, Lee' s thesis is not that 2027 will be strong simply because the calendar changes. Rather, he believes that by then the current drags&mdash AI investment concerns, Federal Reserve uncertainty, deleveraging, and inflation fears&mdash will have largely passed, allowing stronger earnings growth to become the dominant force driving equity markets.

Useful To Me Not Useful To Me

chartiskao
Supreme

05-Aug-2026 15:59

chartiskao About Contact Quote this Post!

x 0
x 0

Alert Admin

https://www.youtube.com/watch?v=2o5suWtdTRM

Executive Summary

Overall Rating: Very Strong (9.5/10)
Tencent is demonstrating three things simultaneously:

Strong double-digit revenue growth.
Expanding profit margins despite heavy AI investment.
Ability to fund AI expansion entirely from operating cash flow.

This combination is uncommon. Many AI companies are sacrificing profits to invest, whereas Tencent is growing both earnings and investment capacity.

1. Revenue Growth (15%)

Revenue reached RMB184.5 billion, up 15% YoY.
That is impressive because Tencent is already one of China' s largest companies.
For comparison:

5% growth is respectable for a mega-cap.
10% is considered strong.
15% is exceptional at Tencent' s scale.

The growth came from multiple businesses rather than one-off gains.

Business	Growth
Gaming	Strong
Advertising	Strong
FinTech	Stable
Cloud	Improving
AI	Beginning monetization

This makes Tencent less dependent on any single business.

2. Operating Leverage Is Improving

3. Core Business Growing 20%

Management highlighted:

Excluding investment gains and losses, the operating businesses grew around 20%.

That means:

WeChat
Advertising
FinTech
Games

are becoming even stronger.
This is healthier than relying on investment portfolio gains.

4. Gaming Remains a Cash Machine

Gaming still contributes roughly half of revenue.

Domestic Games

Games like

Honor of Kings

continue generating enormous cash flow.
These " evergreen" titles are similar to recurring subscription businesses.

International Gaming

Revenue rose:
35%
This is especially encouraging.
Tencent owns or controls stakes in many international studios.
Growth overseas reduces dependence on China' s domestic market.

5. WeChat Is Still Tencent' s Biggest Competitive Advantage

WeChat now has:
1.4 billion monthly active users
Very few companies possess such a platform.
People use WeChat for:

Messaging
Payments
Shopping
Mini Programs
Government services
Banking
Transport
Entertainment

This creates an ecosystem competitors struggle to replicate.

6. AI Is Already Generating Revenue

Many companies are spending heavily on AI without clear returns.
Tencent appears further along.
Examples include:

Advertising

Revenue:
+20%
Management credits AI for:

Better ad targeting
Better recommendation engines
Improved creative generation

This directly increases advertiser ROI.

Customer Service

AI handles merchant inquiries.
Benefits:

Lower labour costs
Faster responses
Better customer satisfaction

Gaming

AI helps developers:

Design characters
Generate dialogue
Test software
Create content

This reduces production costs and shortens development cycles.

7. Massive AI Investment

Capex:
RMB19.1 billion
Up:
120%
That' s enormous.
The spending includes:

NVIDIA GPUs
AI servers
Data centres
Networking equipment

Tencent is investing before AI demand peaks.

8. Hiring AI Talent

Share-based compensation increased nearly:
60%
High-end AI researchers are extremely scarce.
Tencent is competing with:

Alibaba Group
ByteDance
Baidu
DeepSeek

for elite engineers.
This investment could strengthen Tencent' s competitive position over the coming years.

9. Why Margins Matter

Usually companies investing heavily in AI see margins fall.
Tencent achieved:
Higher investment
AND
Higher margins.
That suggests:

Excellent cost control.
Strong pricing power.
High-quality businesses funding AI expansion.

10. Comparison with Global AI Leaders

Company	AI Spending	Current Profitability
Microsoft	Very High	Very High
Meta Platforms	Very High	Very High
Amazon	Very High	Moderate
Tencent	Very High	Very High

Tencent resembles Microsoft and Meta more than companies still searching for profitable AI business models.

Risks

Investors should still monitor:

Slower Chinese consumer spending.
Regulatory changes affecting internet platforms.
U.S. export controls on advanced AI chips.
Competition in AI models and applications.
Gaming approval timelines.

These factors could affect growth or margins.

Long-Term Outlook (2026&ndash 2030)

WeChat' s massive ecosystem,
leading gaming franchises,
growing advertising business,
fintech platform,
and sustained AI investment

chartiskao ( Date: 05-Aug-2026 15:56) Posted:

Why investors pay attention to Tom Lee

Lee has built a reputation because:
Successful calls

Bullish after the COVID-19 crash in 2020.
Correctly anticipated the strong U.S. bull market in 2023&ndash 2025 while many strategists remained cautious.
Has consistently argued that AI would drive a multi-year earnings boom.

Why does he think 2027 could be an outstanding year?

Based on the comments you summarized, Lee believes several temporary headwinds should disappear by then.

1. AI spending begins generating profits

Today, markets worry that companies are spending hundreds of billions on AI infrastructure.
Examples include:

Microsoft
Amazon
Meta Platforms
Alphabet

Lee believes:
2025&ndash 2026

Massive capital expenditure.
Investors worry about returns.

2027

AI applications begin producing significant revenue.
Corporate earnings accelerate.
Investors reward companies that invested early.

This is similar to how internet investments in the late 1990s eventually produced large profits years later.

2. Fed uncertainty disappears

Currently markets are debating:

How much the Federal Reserve will cut rates.
Whether inflation stays under control.
Whether economic growth slows.

Lee expects by 2027:

Inflation is largely controlled.
Interest rates become more predictable.
Companies can plan investments with greater confidence.

Lower uncertainty usually supports higher stock valuations.

3. Housing inflation cools

Housing makes up a large part of U.S. inflation.
Lee points to:

Falling home prices in some regions.
Slower rent increases.
Softer wage growth.

If those trends continue:

Overall inflation declines.
The Fed has more flexibility to ease policy.
Stocks often respond positively.

4. Forced selling is temporary

Lee mentioned several market disruptions.
These include:

AI-related profit taking.
Leveraged investors being forced to sell.
Deleveraging by hedge funds.
Volatility linked to South Korea and semiconductor stocks.

When leverage is unwound:

Selling pressure eventually ends.
Quality companies can recover.
New buyers return.

He refers to this as the market being like a " coiled spring."

5. Strong earnings cycle

Lee expects:

AI productivity to improve profits.
Economic growth to remain positive.
S& P 500 earnings to keep rising.

Historically, stock prices follow earnings over long periods.
If earnings grow 10&ndash 15% annually, indexes can continue reaching new highs.

Why did semiconductor stocks fall?

Lee attributes much of the decline to forced liquidation, not deteriorating business fundamentals.
He specifically mentioned:

Samsung Electronics
SK Hynix

The idea is that leveraged investors sold these holdings to reduce risk, creating sharp short-term declines even if long-term demand for AI chips remained intact.

Is Tom Lee' s view reasonable?

There is logic behind his argument:
Bullish factors

AI investment eventually translates into earnings.
Lower inflation can support lower interest rates.
Reduced policy uncertainty can improve investor confidence.
Corporate profits remain the primary long-term driver of stock prices.

However, there are meaningful risks:

AI spending may not generate the expected returns.
A recession could reduce earnings.
Geopolitical tensions (for example, U.S.&ndash China relations) could disrupt technology supply chains.
Inflation could reaccelerate, limiting the Fed' s ability to ease.

What does this mean for your investment approach?

Given your focus on long-term investing, dividends, and value, Lee' s outlook broadly aligns with a patient strategy:

If AI drives a sustained earnings cycle, large technology companies and semiconductor firms could benefit over the next several years.
Lower interest rates would also tend to support banks, REITs, property developers, and insurers&mdash sectors you' ve been following.
At the same time, maintaining diversification remains important because Lee' s 2027 scenario depends on several macroeconomic assumptions that may not all materialize.

chartiskao ( Date: 05-Aug-2026 09:54) Posted:

https://www.youtube.com/watch?v=bYokg98l2tI& t=6s

The song is not simply about being " too kind." It is about the painful realization that love cannot be sustained by one person' s sacrifice alone. The narrator gradually understands that loyalty, patience, and forgiveness cannot make someone love you back if their feelings are no longer there.
At around 3:23, the emotional turning point becomes clearer. Rather than pleading for the relationship to continue, the narrator begins to accept a difficult truth:

Continuing to wait for someone who does not appreciate your love is a form of self-deception.
Mistaking endurance for love only prolongs the pain.
The relationship has become one-sided, where one person gives while the other has emotionally moved on.

The word " delusion" (or " illusion" ) is significant because it suggests the narrator recognizes they were holding onto hope rather than reality. They were not only waiting for the other person to change&mdash they were also convincing themselves that unconditional devotion would eventually be rewarded.

Emotional progression

Devotion &ndash The narrator loves wholeheartedly and overlooks the partner' s shortcomings.
Tolerance &ndash They continue forgiving hurtful behavior, believing perseverance will save the relationship.
Disillusionment &ndash They realize their efforts are not changing anything.
Acceptance &ndash They understand that remaining in the relationship is based on illusion rather than mutual love.

The deeper message

The song is less about blaming the unfaithful partner and more about encouraging self-awareness. It suggests that:

Love should be reciprocal.
Kindness is a virtue, but it should not come at the cost of self-respect.
Sometimes the hardest act of love is knowing when to let go rather than continuing to endure emotional pain.

This is why the ballad resonates with many listeners: it reflects a universal experience of discovering that there is a difference between loving someone deeply and losing yourself trying to keep someone who no longer values that love.

Useful To Me Not Useful To Me

chartiskao
Supreme

05-Aug-2026 15:56

chartiskao About Contact Quote this Post!

x 0
x 0

Alert Admin

Why investors pay attention to Tom Lee

Lee has built a reputation because:
Successful calls

Bullish after the COVID-19 crash in 2020.
Correctly anticipated the strong U.S. bull market in 2023&ndash 2025 while many strategists remained cautious.
Has consistently argued that AI would drive a multi-year earnings boom.

Why does he think 2027 could be an outstanding year?

Based on the comments you summarized, Lee believes several temporary headwinds should disappear by then.

1. AI spending begins generating profits

Today, markets worry that companies are spending hundreds of billions on AI infrastructure.
Examples include:

Microsoft
Amazon
Meta Platforms
Alphabet

Lee believes:
2025&ndash 2026

Massive capital expenditure.
Investors worry about returns.

2027

AI applications begin producing significant revenue.
Corporate earnings accelerate.
Investors reward companies that invested early.

This is similar to how internet investments in the late 1990s eventually produced large profits years later.

2. Fed uncertainty disappears

Currently markets are debating:

How much the Federal Reserve will cut rates.
Whether inflation stays under control.
Whether economic growth slows.

Lee expects by 2027:

Inflation is largely controlled.
Interest rates become more predictable.
Companies can plan investments with greater confidence.

Lower uncertainty usually supports higher stock valuations.

3. Housing inflation cools

Housing makes up a large part of U.S. inflation.
Lee points to:

Falling home prices in some regions.
Slower rent increases.
Softer wage growth.

If those trends continue:

Overall inflation declines.
The Fed has more flexibility to ease policy.
Stocks often respond positively.

4. Forced selling is temporary

Lee mentioned several market disruptions.
These include:

AI-related profit taking.
Leveraged investors being forced to sell.
Deleveraging by hedge funds.
Volatility linked to South Korea and semiconductor stocks.

When leverage is unwound:

Selling pressure eventually ends.
Quality companies can recover.
New buyers return.

He refers to this as the market being like a " coiled spring."

5. Strong earnings cycle

Lee expects:

AI productivity to improve profits.
Economic growth to remain positive.
S& P 500 earnings to keep rising.

Historically, stock prices follow earnings over long periods.
If earnings grow 10&ndash 15% annually, indexes can continue reaching new highs.

Why did semiconductor stocks fall?

Lee attributes much of the decline to forced liquidation, not deteriorating business fundamentals.
He specifically mentioned:

Samsung Electronics
SK Hynix

The idea is that leveraged investors sold these holdings to reduce risk, creating sharp short-term declines even if long-term demand for AI chips remained intact.

Is Tom Lee' s view reasonable?

There is logic behind his argument:
Bullish factors

AI investment eventually translates into earnings.
Lower inflation can support lower interest rates.
Reduced policy uncertainty can improve investor confidence.
Corporate profits remain the primary long-term driver of stock prices.

However, there are meaningful risks:

AI spending may not generate the expected returns.
A recession could reduce earnings.
Geopolitical tensions (for example, U.S.&ndash China relations) could disrupt technology supply chains.
Inflation could reaccelerate, limiting the Fed' s ability to ease.

What does this mean for your investment approach?

Given your focus on long-term investing, dividends, and value, Lee' s outlook broadly aligns with a patient strategy:

If AI drives a sustained earnings cycle, large technology companies and semiconductor firms could benefit over the next several years.
Lower interest rates would also tend to support banks, REITs, property developers, and insurers&mdash sectors you' ve been following.
At the same time, maintaining diversification remains important because Lee' s 2027 scenario depends on several macroeconomic assumptions that may not all materialize.

chartiskao ( Date: 05-Aug-2026 09:54) Posted:

Continuing to wait for someone who does not appreciate your love is a form of self-deception.
Mistaking endurance for love only prolongs the pain.
The relationship has become one-sided, where one person gives while the other has emotionally moved on.

Emotional progression

Devotion &ndash The narrator loves wholeheartedly and overlooks the partner' s shortcomings.
Tolerance &ndash They continue forgiving hurtful behavior, believing perseverance will save the relationship.
Disillusionment &ndash They realize their efforts are not changing anything.
Acceptance &ndash They understand that remaining in the relationship is based on illusion rather than mutual love.

The deeper message

The song is less about blaming the unfaithful partner and more about encouraging self-awareness. It suggests that:

Love should be reciprocal.
Kindness is a virtue, but it should not come at the cost of self-respect.
Sometimes the hardest act of love is knowing when to let go rather than continuing to endure emotional pain.

chartiskao ( Date: 05-Aug-2026 09:50) Posted:

https://www.youtube.com/watch?v=bYokg98l2tI& t=6s

Large institutional investors (" big funds" ) do not usually target Tencent because it is a bad company. Instead, they often target it because it is one of the largest, most liquid stocks in Asia, making it an efficient vehicle for adjusting risk when market sentiment changes.

Why big funds sell Tencent first

1. Liquidity

Tencent trades billions of Hong Kong dollars every day.
A fund managing US$50 billion cannot easily sell a small company without moving the price significantly. Tencent, however, allows them to buy or sell hundreds of millions of dollars' worth of shares relatively quickly.
Think of it like this:

Small stock = narrow door
Tencent = six-lane highway

When funds need cash, they use the highway.

2. Portfolio rebalancing

During an AI correction, many global funds reduce exposure to technology.
Suppose a fund owns:

Nvidia
Microsoft
Amazon
Alphabet
Meta
Tencent
TSMC

If the investment committee decides, " Reduce AI exposure by 15%," Tencent is one of the easiest positions to trim because it is a large, liquid holding.
The sale is often driven by portfolio strategy, not a judgment that Tencent' s business has suddenly deteriorated.

3. Risk management

Large funds operate under risk limits.
If technology stocks become more volatile:

Value-at-Risk (VaR) increases.
Portfolio volatility rises.
Risk managers may require portfolio managers to cut exposure.

The fastest way is to sell large liquid holdings such as Tencent.

4. ETF and index flows

Tencent is a major component of:

Hang Seng Index
Hang Seng Tech Index
MSCI China
FTSE China
Many Asia and emerging-market ETFs

When investors withdraw money from these funds:
Investor redemption &rarr ETF manager sells Tencent automatically.
The selling is mechanical rather than based on Tencent-specific news.

5. Hedge funds use Tencent as a trading vehicle

Many hedge funds trade Tencent because it is highly liquid.
Examples include:

Long Tencent, short Alibaba
Long Tencent, short Nasdaq
Long Tencent, short Hang Seng futures

When these trades are unwound, Tencent may experience heavy selling even if there is no negative company news.

Market psychology: the roller coaster

The bigger story is often psychological rather than fundamental.

Stage 1 &ndash Excitement

" AI will transform everything."
Funds buy aggressively.
Prices rise rapidly.
Fear of missing out (FOMO) brings in more buyers.
Momentum feeds on itself.

Stage 2 &ndash Euphoria

Analysts continually raise price targets.
Retail investors rush in.
Valuations become stretched.
Almost any AI-related announcement is rewarded.
People begin to believe prices can only go higher.

Stage 3 &ndash Doubt

Small disappointments appear:

Higher AI spending
Slower monetisation
Margin pressure

Investors begin asking:

" Are we paying too much?"

The first wave of selling starts.

Stage 4 &ndash Panic

This is where psychology changes dramatically.
Investors think:

" What if this becomes another dot-com bubble?"
" Protect profits."
" Sell before everyone else."

Selling accelerates.
Algorithms detect downward momentum and sell.
Margin calls force leveraged investors to sell.
ETFs experience redemptions.
Each wave of selling reinforces the next.

Stage 5 &ndash Capitulation

This is often the most emotional stage.
News headlines become overwhelmingly negative.
Financial television focuses on falling prices.
Retail investors lose confidence and sell near the lows.
Ironically, this is often when many long-term investors begin buying.

Stage 6 &ndash Recovery

Eventually, fundamentals become more important than fear.
Investors recognise that companies like Tencent still have:

billions of users,
strong cash flow,
profitable businesses,
AI capabilities that can generate future earnings.

Confidence gradually returns.

Why quality companies can fall sharply

Imagine Tencent earns HK$100 billion annually.
One week:

Nothing changes in its business.
WeChat still has its users.
Games are still generating revenue.
Cloud services continue operating.

Yet the share price falls 15%.
Why?
Because the market is a marketplace, not a valuation machine in the short run. Prices reflect what buyers and sellers are willing to pay today, influenced by fear, optimism, liquidity needs, and risk management&mdash not just intrinsic value.
This idea aligns with Benjamin Graham' s famous analogy of " Mr. Market" : the market' s mood swings can cause prices to fluctuate much more than a company' s underlying value. Long-term investors aim to distinguish between changes in price and changes in business value.
During major thematic corrections&mdash whether the dot-com bust, the 2022 technology sell-off, or the 2026 AI reset&mdash high-quality companies like Tencent can experience large declines because they are large, liquid, widely owned, and central to investors' portfolios. That combination makes them a common source of funds when institutions need to reduce risk quickly.

Useful To Me Not Useful To Me

chartiskao
Supreme

05-Aug-2026 09:54

chartiskao About Contact Quote this Post!

x 0
x 0

Alert Admin

Continuing to wait for someone who does not appreciate your love is a form of self-deception.
Mistaking endurance for love only prolongs the pain.
The relationship has become one-sided, where one person gives while the other has emotionally moved on.

Emotional progression

Devotion &ndash The narrator loves wholeheartedly and overlooks the partner' s shortcomings.
Tolerance &ndash They continue forgiving hurtful behavior, believing perseverance will save the relationship.
Disillusionment &ndash They realize their efforts are not changing anything.
Acceptance &ndash They understand that remaining in the relationship is based on illusion rather than mutual love.

The deeper message

The song is less about blaming the unfaithful partner and more about encouraging self-awareness. It suggests that:

Love should be reciprocal.
Kindness is a virtue, but it should not come at the cost of self-respect.
Sometimes the hardest act of love is knowing when to let go rather than continuing to endure emotional pain.

chartiskao ( Date: 05-Aug-2026 09:50) Posted:

Why big funds sell Tencent first

1. Liquidity

Small stock = narrow door
Tencent = six-lane highway

When funds need cash, they use the highway.

2. Portfolio rebalancing

During an AI correction, many global funds reduce exposure to technology.
Suppose a fund owns:

Nvidia
Microsoft
Amazon
Alphabet
Meta
Tencent
TSMC

3. Risk management

Large funds operate under risk limits.
If technology stocks become more volatile:

Value-at-Risk (VaR) increases.
Portfolio volatility rises.
Risk managers may require portfolio managers to cut exposure.

The fastest way is to sell large liquid holdings such as Tencent.

4. ETF and index flows

Tencent is a major component of:

Hang Seng Index
Hang Seng Tech Index
MSCI China
FTSE China
Many Asia and emerging-market ETFs

When investors withdraw money from these funds:
Investor redemption &rarr ETF manager sells Tencent automatically.
The selling is mechanical rather than based on Tencent-specific news.

5. Hedge funds use Tencent as a trading vehicle

Many hedge funds trade Tencent because it is highly liquid.
Examples include:

Long Tencent, short Alibaba
Long Tencent, short Nasdaq
Long Tencent, short Hang Seng futures

When these trades are unwound, Tencent may experience heavy selling even if there is no negative company news.

Market psychology: the roller coaster

The bigger story is often psychological rather than fundamental.

Stage 1 &ndash Excitement

" AI will transform everything."
Funds buy aggressively.
Prices rise rapidly.
Fear of missing out (FOMO) brings in more buyers.
Momentum feeds on itself.

Stage 2 &ndash Euphoria

Analysts continually raise price targets.
Retail investors rush in.
Valuations become stretched.
Almost any AI-related announcement is rewarded.
People begin to believe prices can only go higher.

Stage 3 &ndash Doubt

Small disappointments appear:

Higher AI spending
Slower monetisation
Margin pressure

Investors begin asking:

" Are we paying too much?"

The first wave of selling starts.

Stage 4 &ndash Panic

This is where psychology changes dramatically.
Investors think:

" What if this becomes another dot-com bubble?"
" Protect profits."
" Sell before everyone else."

Selling accelerates.
Algorithms detect downward momentum and sell.
Margin calls force leveraged investors to sell.
ETFs experience redemptions.
Each wave of selling reinforces the next.

Stage 5 &ndash Capitulation

Stage 6 &ndash Recovery

Eventually, fundamentals become more important than fear.
Investors recognise that companies like Tencent still have:

billions of users,
strong cash flow,
profitable businesses,
AI capabilities that can generate future earnings.

Confidence gradually returns.

Why quality companies can fall sharply

Imagine Tencent earns HK$100 billion annually.
One week:

Nothing changes in its business.
WeChat still has its users.
Games are still generating revenue.
Cloud services continue operating.

chartiskao ( Date: 05-Aug-2026 09:46) Posted:

the mood of what many investors called the 2026 global AI washout. Tencent (HKEX: 0700) was caught in the broad sell-off even though its underlying business remained profitable.

Tencent during the AI washout (5 August 2026)

From your screenshot:

Share price: HK$482.80
Down: HK$4.80 (-0.98%)
Day' s range: HK$482.20 &ndash HK$493.40
Market capitalization: HK$4.49 trillion
P/E: approximately 17.3×

Although the decline on this day was less than 1%, it came after several weeks of heavy selling across AI and technology stocks.

Why Tencent became a " roller coaster"

The volatility was driven by several factors occurring at the same time:

Factor	Impact on Tencent
Global AI valuation reset	Investors questioned whether massive AI spending would generate sufficient returns.
Gaming concerns	Data pointing to softer mobile gaming revenue raised worries ahead of quarterly earnings.
Rotation out of technology	Money flowed from expensive AI-related stocks into more defensive sectors.
China technology weakness	The Hang Seng Tech Index and many Chinese technology stocks experienced broad selling pressure.

Yet Tencent' s fundamentals remained resilient

Despite the share price swings, Tencent continued to report:

Revenue growth
Double-digit operating profit growth
Continued investment in AI infrastructure
Strong cash generation
Expansion of AI across WeChat, advertising, cloud computing and enterprise software.

This highlights an important distinction:

Stock price reflects investor sentiment, which can change rapidly.
Business value changes much more gradually.

The broader 2026 AI washout

The sell-off was not limited to Tencent.
Across global markets, investors reduced exposure to:

AI software companies
Semiconductor manufacturers
Data-centre suppliers
Cloud infrastructure companies
Chinese technology stocks

Several Asian markets experienced sharp declines as investors reassessed AI valuations and future earnings expectations.

Historical perspective

Tencent has experienced several major drawdowns over the past decade:

2018: Chinese gaming regulation
2021&ndash 2022: Internet regulatory crackdown
2026: Global AI valuation reset

Each episode produced substantial volatility, but the company' s long-term earnings power depended on the performance of its core businesses rather than short-term market sentiment.
Your screenshot is therefore a good illustration of how market psychology can resemble a roller coaster during major thematic corrections. Even one of Asia' s highest-quality technology companies experienced significant swings as investors rapidly repriced AI-related assets before focusing again on long-term fundamentals.

Useful To Me Not Useful To Me

chartiskao
Supreme

05-Aug-2026 09:50

chartiskao About Contact Quote this Post!

x 0
x 0

Alert Admin

Why big funds sell Tencent first

1. Liquidity

Small stock = narrow door
Tencent = six-lane highway

When funds need cash, they use the highway.

2. Portfolio rebalancing

During an AI correction, many global funds reduce exposure to technology.
Suppose a fund owns:

Nvidia
Microsoft
Amazon
Alphabet
Meta
Tencent
TSMC

3. Risk management

Large funds operate under risk limits.
If technology stocks become more volatile:

Value-at-Risk (VaR) increases.
Portfolio volatility rises.
Risk managers may require portfolio managers to cut exposure.

The fastest way is to sell large liquid holdings such as Tencent.

4. ETF and index flows

Tencent is a major component of:

Hang Seng Index
Hang Seng Tech Index
MSCI China
FTSE China
Many Asia and emerging-market ETFs

When investors withdraw money from these funds:
Investor redemption &rarr ETF manager sells Tencent automatically.
The selling is mechanical rather than based on Tencent-specific news.

5. Hedge funds use Tencent as a trading vehicle

Many hedge funds trade Tencent because it is highly liquid.
Examples include:

Long Tencent, short Alibaba
Long Tencent, short Nasdaq
Long Tencent, short Hang Seng futures

When these trades are unwound, Tencent may experience heavy selling even if there is no negative company news.

Market psychology: the roller coaster

The bigger story is often psychological rather than fundamental.

Stage 1 &ndash Excitement

" AI will transform everything."
Funds buy aggressively.
Prices rise rapidly.
Fear of missing out (FOMO) brings in more buyers.
Momentum feeds on itself.

Stage 2 &ndash Euphoria

Analysts continually raise price targets.
Retail investors rush in.
Valuations become stretched.
Almost any AI-related announcement is rewarded.
People begin to believe prices can only go higher.

Stage 3 &ndash Doubt

Small disappointments appear:

Higher AI spending
Slower monetisation
Margin pressure

Investors begin asking:

" Are we paying too much?"

The first wave of selling starts.

Stage 4 &ndash Panic

This is where psychology changes dramatically.
Investors think:

" What if this becomes another dot-com bubble?"
" Protect profits."
" Sell before everyone else."

Selling accelerates.
Algorithms detect downward momentum and sell.
Margin calls force leveraged investors to sell.
ETFs experience redemptions.
Each wave of selling reinforces the next.

Stage 5 &ndash Capitulation

Stage 6 &ndash Recovery

Eventually, fundamentals become more important than fear.
Investors recognise that companies like Tencent still have:

billions of users,
strong cash flow,
profitable businesses,
AI capabilities that can generate future earnings.

Confidence gradually returns.

Why quality companies can fall sharply

Imagine Tencent earns HK$100 billion annually.
One week:

Nothing changes in its business.
WeChat still has its users.
Games are still generating revenue.
Cloud services continue operating.

chartiskao ( Date: 05-Aug-2026 09:46) Posted:

the mood of what many investors called the 2026 global AI washout. Tencent (HKEX: 0700) was caught in the broad sell-off even though its underlying business remained profitable.

Tencent during the AI washout (5 August 2026)

From your screenshot:

Share price: HK$482.80
Down: HK$4.80 (-0.98%)
Day' s range: HK$482.20 &ndash HK$493.40
Market capitalization: HK$4.49 trillion
P/E: approximately 17.3×

Although the decline on this day was less than 1%, it came after several weeks of heavy selling across AI and technology stocks.

Why Tencent became a " roller coaster"

The volatility was driven by several factors occurring at the same time:

Factor	Impact on Tencent
Global AI valuation reset	Investors questioned whether massive AI spending would generate sufficient returns.
Gaming concerns	Data pointing to softer mobile gaming revenue raised worries ahead of quarterly earnings.
Rotation out of technology	Money flowed from expensive AI-related stocks into more defensive sectors.
China technology weakness	The Hang Seng Tech Index and many Chinese technology stocks experienced broad selling pressure.

Yet Tencent' s fundamentals remained resilient

Despite the share price swings, Tencent continued to report:

Revenue growth
Double-digit operating profit growth
Continued investment in AI infrastructure
Strong cash generation
Expansion of AI across WeChat, advertising, cloud computing and enterprise software.

This highlights an important distinction:

Stock price reflects investor sentiment, which can change rapidly.
Business value changes much more gradually.

The broader 2026 AI washout

The sell-off was not limited to Tencent.
Across global markets, investors reduced exposure to:

AI software companies
Semiconductor manufacturers
Data-centre suppliers
Cloud infrastructure companies
Chinese technology stocks

Several Asian markets experienced sharp declines as investors reassessed AI valuations and future earnings expectations.

Historical perspective

Tencent has experienced several major drawdowns over the past decade:

2018: Chinese gaming regulation
2021&ndash 2022: Internet regulatory crackdown
2026: Global AI valuation reset

chartiskao ( Date: 05-Aug-2026 08:42) Posted:

This week (early August 2026) has been characterized by a sharp rotation within AI-related sectors, rather than indiscriminate selling. Investors are moving away from highly valued AI infrastructure names and becoming much more selective.

1. AI infrastructure & semiconductor stocks: Heavy selling

The biggest losses have been in AI chip and infrastructure stocks.
Reasons:

Investors are questioning whether the enormous AI capital spending will generate sufficient returns.
Hedge funds heavily exposed to AI unwound positions after July' s sell-off.
Valuation concerns remain elevated despite strong earnings.

Examples of pressure

Nvidia
AMD
Broadcom
Marvell
CoreWeave
Memory-related AI suppliers

These companies remain fundamentally strong but are experiencing multiple compression as investors reassess valuations.

2. AI software: Mixed performance

Earlier in 2026, AI software companies suffered a historic sell-off as investors feared that increasingly capable AI agents could disrupt traditional SaaS business models.
Large software companies such as:

Microsoft
Salesforce
ServiceNow

experienced significant declines during that correction.
However, this week the picture has improved.
Strong cloud earnings from Microsoft, Google, and Amazon have renewed confidence that enterprise AI adoption continues to accelerate, helping software stocks rebound.

3. AI cybersecurity: One of the strongest themes

Cybersecurity continues to be viewed as one of the more resilient AI sectors.
Why?
As companies deploy AI:

cyber attacks become more sophisticated,
AI agents require identity protection,
AI workloads need new security layers,
compliance requirements increase.

This has kept investor interest relatively strong.
Companies benefiting include:

Palo Alto Networks
CrowdStrike
Zscaler
Fortinet

Many analysts now see cybersecurity as a more durable AI investment theme than some pure AI software names.

Market rotation this week

Sector	This week	Outlook
AI Chips	⭐ Weak	Valuation reset continues
AI Infrastructure	⭐ Weak	Investors reducing leverage
AI Software	Neutral to Positive	Supported by strong cloud earnings
AI Cybersecurity	Positive	Structural growth story remains intact
Cloud Providers	Strong	AI spending still increasing

What Wall Street is watching

The market is now focusing on three questions:

Can hyperscalers earn attractive returns on AI investments?
Will enterprise customers spend more on AI software?
Can cybersecurity companies monetize AI-driven security demand?

If these answers remain positive, many strategists believe the recent sell-off represents a healthy reset rather than the end of the AI investment cycle. Others, including valuation experts, continue to warn that smaller AI companies may face further downside if earnings fail to justify lofty expectations.
For a value investor, the current environment suggests:

Be cautious with highly valued AI infrastructure stocks.
Favor profitable companies with recurring revenue and strong cash flow.
AI cybersecurity appears to offer one of the better long-term risk/reward profiles, as demand is supported by the increasing need to secure AI systems rather than by speculative AI adoption alone.

Useful To Me Not Useful To Me

chartiskao
Supreme

05-Aug-2026 09:46

chartiskao About Contact Quote this Post!

x 0
x 0

Alert Admin

the mood of what many investors called the 2026 global AI washout. Tencent (HKEX: 0700) was caught in the broad sell-off even though its underlying business remained profitable.

Tencent during the AI washout (5 August 2026)

From your screenshot:

Share price: HK$482.80
Down: HK$4.80 (-0.98%)
Day' s range: HK$482.20 &ndash HK$493.40
Market capitalization: HK$4.49 trillion
P/E: approximately 17.3×

Although the decline on this day was less than 1%, it came after several weeks of heavy selling across AI and technology stocks.

Why Tencent became a " roller coaster"

The volatility was driven by several factors occurring at the same time:

Factor	Impact on Tencent
Global AI valuation reset	Investors questioned whether massive AI spending would generate sufficient returns.
Gaming concerns	Data pointing to softer mobile gaming revenue raised worries ahead of quarterly earnings.
Rotation out of technology	Money flowed from expensive AI-related stocks into more defensive sectors.
China technology weakness	The Hang Seng Tech Index and many Chinese technology stocks experienced broad selling pressure.

Yet Tencent' s fundamentals remained resilient

Despite the share price swings, Tencent continued to report:

Revenue growth
Double-digit operating profit growth
Continued investment in AI infrastructure
Strong cash generation
Expansion of AI across WeChat, advertising, cloud computing and enterprise software.

This highlights an important distinction:

Stock price reflects investor sentiment, which can change rapidly.
Business value changes much more gradually.

The broader 2026 AI washout

The sell-off was not limited to Tencent.
Across global markets, investors reduced exposure to:

AI software companies
Semiconductor manufacturers
Data-centre suppliers
Cloud infrastructure companies
Chinese technology stocks

Several Asian markets experienced sharp declines as investors reassessed AI valuations and future earnings expectations.

Historical perspective

Tencent has experienced several major drawdowns over the past decade:

2018: Chinese gaming regulation
2021&ndash 2022: Internet regulatory crackdown
2026: Global AI valuation reset

chartiskao ( Date: 05-Aug-2026 08:42) Posted:

1. AI infrastructure & semiconductor stocks: Heavy selling

The biggest losses have been in AI chip and infrastructure stocks.
Reasons:

Investors are questioning whether the enormous AI capital spending will generate sufficient returns.
Hedge funds heavily exposed to AI unwound positions after July' s sell-off.
Valuation concerns remain elevated despite strong earnings.

Examples of pressure

Nvidia
AMD
Broadcom
Marvell
CoreWeave
Memory-related AI suppliers

These companies remain fundamentally strong but are experiencing multiple compression as investors reassess valuations.

2. AI software: Mixed performance

Microsoft
Salesforce
ServiceNow

3. AI cybersecurity: One of the strongest themes

Cybersecurity continues to be viewed as one of the more resilient AI sectors.
Why?
As companies deploy AI:

cyber attacks become more sophisticated,
AI agents require identity protection,
AI workloads need new security layers,
compliance requirements increase.

This has kept investor interest relatively strong.
Companies benefiting include:

Palo Alto Networks
CrowdStrike
Zscaler
Fortinet

Many analysts now see cybersecurity as a more durable AI investment theme than some pure AI software names.

Market rotation this week

Sector	This week	Outlook
AI Chips	⭐ Weak	Valuation reset continues
AI Infrastructure	⭐ Weak	Investors reducing leverage
AI Software	Neutral to Positive	Supported by strong cloud earnings
AI Cybersecurity	Positive	Structural growth story remains intact
Cloud Providers	Strong	AI spending still increasing

What Wall Street is watching

The market is now focusing on three questions:

Can hyperscalers earn attractive returns on AI investments?
Will enterprise customers spend more on AI software?
Can cybersecurity companies monetize AI-driven security demand?

Be cautious with highly valued AI infrastructure stocks.
Favor profitable companies with recurring revenue and strong cash flow.
AI cybersecurity appears to offer one of the better long-term risk/reward profiles, as demand is supported by the increasing need to secure AI systems rather than by speculative AI adoption alone.

chartiskao ( Date: 04-Aug-2026 20:27) Posted:

https://www.youtube.com/watch?v=yoy-o-lccKE

Here' s a concise strategic summary of the history of Google based on your notes.

The Evolution of Google: From University Research to Global Technology Leader

1. The Early Days of Search (1990s)

Before Google, internet search was inefficient.

Search engines such as AltaVista, Lycos, and Excite mainly ranked websites by counting how often keywords appeared.
This encouraged website owners to " stuff" pages with repeated keywords to manipulate rankings.
Search results were often inaccurate, irrelevant, and difficult for users.

At the same time, Yahoo dominated the internet through its portal strategy.
Yahoo believed users should:

Stay on Yahoo' s homepage.
Read Yahoo News, Finance, Sports, and Email.
Generate advertising revenue by spending more time within Yahoo' s ecosystem.

Searching was only one feature among many.

2. The Birth of Google (1996&ndash 1998)

At Stanford University, PhD students Larry Page and Sergey Brin developed a new search algorithm called PageRank.
Instead of simply counting keywords, PageRank viewed hyperlinks as recommendations.
The idea was simple:

Every website link acted as a " vote."
Votes from trusted websites carried greater weight.
Pages receiving many high-quality links ranked higher.

This approach measured authority rather than keyword repetition.
The result was dramatically better search quality.

3. The Missed Opportunity

Despite building a superior search engine, Page and Brin initially wanted to sell it.
They approached Yahoo and later Excite.
Their asking price was only US$1 million.
Both companies rejected the offer.

Why?

Their business philosophies were fundamentally different.

Yahoo' s philosophy

Keep users inside Yahoo.
Increase page views.
Display more advertisements.

Google' s philosophy

Help users find exactly what they need as quickly as possible.
Let users leave Google immediately after getting useful results.

At the time, many executives believed Google' s approach would reduce advertising opportunities because users spent less time on the website.
Ironically, this became Google' s greatest strength.

4. Google' s Early Growth

A major turning point came when Andy Bechtolsheim was shown the search engine.
Impressed within minutes, he wrote a cheque for US$100,000 before Google was even officially incorporated.
Google Inc. was incorporated in 1998.
This funding allowed the founders to leave academia and build the company full-time.

5. The Revolutionary Revenue Model

Google initially had excellent technology but no business model.
Everything changed in 2000 with Google AdWords.
The innovation was the pay-per-click (PPC) advertising model.
Instead of paying simply to display advertisements:

Advertisers only paid when users clicked.
Ads were highly relevant to search queries.
Businesses could accurately measure return on investment.

This created a win-win situation:

Users received relevant advertisements.
Businesses reached interested customers.
Google earned revenue only when advertising created value.

The model proved enormously scalable and became one of the most successful advertising systems in history.

6. Going Public (2004)

Google' s Initial Public Offering (IPO) in 2004 became one of the most important technology listings ever.
Its successful public debut:

Generated billions of dollars.
Demonstrated that internet companies could build profitable businesses.
Helped restore investor confidence after the dot-com crash.

The IPO also provided Google with resources for long-term expansion.

7. Beyond Search

Google evolved from a search engine into a diversified technology company.
Major milestones included:

Acquiring YouTube in 2006, becoming the world' s largest video platform.
Developing the Android operating system, now powering billions of smartphones.
Expanding into cloud computing, artificial intelligence, digital maps, productivity software, autonomous driving research, and hardware.

In 2015, Google reorganized under its parent company, Alphabet Inc., allowing its various businesses to operate more independently.

8. Strategic Lessons

Google' s rise illustrates several important business principles:

Lesson	Explanation
Solve the user' s problem first	Superior user experience built long-term loyalty.
Better technology creates competitive advantage	PageRank produced significantly more relevant search results.
A strong business model amplifies innovation	AdWords transformed excellent technology into a highly profitable enterprise.
Short-term thinking can miss transformative opportunities	Yahoo and Excite underestimated Google' s disruptive potential.
Focus can outperform feature overload	Google concentrated on delivering the best search experience rather than becoming a broad internet portal.

Conclusion

Google' s success was not merely the result of creating a better search engine. It fundamentally changed how information is discovered and monetized on the internet. By prioritizing relevance, speed, and user satisfaction through PageRank, and later introducing the highly effective AdWords pay-per-click model, Google established a durable competitive advantage. The decision by Yahoo and Excite to reject acquiring Google for just US$1 million is now regarded as one of the most significant missed opportunities in technology and business history.

Useful To Me Not Useful To Me

chartiskao
Supreme

05-Aug-2026 08:42

chartiskao About Contact Quote this Post!

x 0
x 0

Alert Admin

1. AI infrastructure & semiconductor stocks: Heavy selling

The biggest losses have been in AI chip and infrastructure stocks.
Reasons:

Investors are questioning whether the enormous AI capital spending will generate sufficient returns.
Hedge funds heavily exposed to AI unwound positions after July' s sell-off.
Valuation concerns remain elevated despite strong earnings.

Examples of pressure

Nvidia
AMD
Broadcom
Marvell
CoreWeave
Memory-related AI suppliers

These companies remain fundamentally strong but are experiencing multiple compression as investors reassess valuations.

2. AI software: Mixed performance

Microsoft
Salesforce
ServiceNow

3. AI cybersecurity: One of the strongest themes

Cybersecurity continues to be viewed as one of the more resilient AI sectors.
Why?
As companies deploy AI:

cyber attacks become more sophisticated,
AI agents require identity protection,
AI workloads need new security layers,
compliance requirements increase.

This has kept investor interest relatively strong.
Companies benefiting include:

Palo Alto Networks
CrowdStrike
Zscaler
Fortinet

Many analysts now see cybersecurity as a more durable AI investment theme than some pure AI software names.

Market rotation this week

Sector	This week	Outlook
AI Chips	⭐ Weak	Valuation reset continues
AI Infrastructure	⭐ Weak	Investors reducing leverage
AI Software	Neutral to Positive	Supported by strong cloud earnings
AI Cybersecurity	Positive	Structural growth story remains intact
Cloud Providers	Strong	AI spending still increasing

What Wall Street is watching

The market is now focusing on three questions:

Can hyperscalers earn attractive returns on AI investments?
Will enterprise customers spend more on AI software?
Can cybersecurity companies monetize AI-driven security demand?

Be cautious with highly valued AI infrastructure stocks.
Favor profitable companies with recurring revenue and strong cash flow.
AI cybersecurity appears to offer one of the better long-term risk/reward profiles, as demand is supported by the increasing need to secure AI systems rather than by speculative AI adoption alone.

chartiskao ( Date: 04-Aug-2026 20:27) Posted:

https://www.youtube.com/watch?v=yoy-o-lccKE

Here' s a concise strategic summary of the history of Google based on your notes.

The Evolution of Google: From University Research to Global Technology Leader

1. The Early Days of Search (1990s)

Before Google, internet search was inefficient.

Search engines such as AltaVista, Lycos, and Excite mainly ranked websites by counting how often keywords appeared.
This encouraged website owners to " stuff" pages with repeated keywords to manipulate rankings.
Search results were often inaccurate, irrelevant, and difficult for users.

At the same time, Yahoo dominated the internet through its portal strategy.
Yahoo believed users should:

Stay on Yahoo' s homepage.
Read Yahoo News, Finance, Sports, and Email.
Generate advertising revenue by spending more time within Yahoo' s ecosystem.

Searching was only one feature among many.

2. The Birth of Google (1996&ndash 1998)

Every website link acted as a " vote."
Votes from trusted websites carried greater weight.
Pages receiving many high-quality links ranked higher.

This approach measured authority rather than keyword repetition.
The result was dramatically better search quality.

3. The Missed Opportunity

Why?

Their business philosophies were fundamentally different.

Yahoo' s philosophy

Keep users inside Yahoo.
Increase page views.
Display more advertisements.

Google' s philosophy

Help users find exactly what they need as quickly as possible.
Let users leave Google immediately after getting useful results.

At the time, many executives believed Google' s approach would reduce advertising opportunities because users spent less time on the website.
Ironically, this became Google' s greatest strength.

4. Google' s Early Growth

5. The Revolutionary Revenue Model

Advertisers only paid when users clicked.
Ads were highly relevant to search queries.
Businesses could accurately measure return on investment.

This created a win-win situation:

Users received relevant advertisements.
Businesses reached interested customers.
Google earned revenue only when advertising created value.

The model proved enormously scalable and became one of the most successful advertising systems in history.

6. Going Public (2004)

Google' s Initial Public Offering (IPO) in 2004 became one of the most important technology listings ever.
Its successful public debut:

Generated billions of dollars.
Demonstrated that internet companies could build profitable businesses.
Helped restore investor confidence after the dot-com crash.

The IPO also provided Google with resources for long-term expansion.

7. Beyond Search

Google evolved from a search engine into a diversified technology company.
Major milestones included:

Acquiring YouTube in 2006, becoming the world' s largest video platform.
Developing the Android operating system, now powering billions of smartphones.
Expanding into cloud computing, artificial intelligence, digital maps, productivity software, autonomous driving research, and hardware.

In 2015, Google reorganized under its parent company, Alphabet Inc., allowing its various businesses to operate more independently.

8. Strategic Lessons

Google' s rise illustrates several important business principles:

Lesson	Explanation
Solve the user' s problem first	Superior user experience built long-term loyalty.
Better technology creates competitive advantage	PageRank produced significantly more relevant search results.
A strong business model amplifies innovation	AdWords transformed excellent technology into a highly profitable enterprise.
Short-term thinking can miss transformative opportunities	Yahoo and Excite underestimated Google' s disruptive potential.
Focus can outperform feature overload	Google concentrated on delivering the best search experience rather than becoming a broad internet portal.

Conclusion

chartiskao ( Date: 04-Aug-2026 14:25) Posted:

I can discuss the themes of the song and apply them as a metaphor to today' s AI bubble, China, SoftBank, and the global financial system.

" Wishful Thinking" as an investment philosophy

The title itself captures one of the biggest risks in financial markets:

Confusing hope with probability.

Every major bubble contains an element of wishful thinking.

1. AI Bubble

Wishful thinking

Investors assume:

AI demand will grow forever.
Every AI company will become profitable.
Capital will remain cheap.
Governments will always support AI.

Reality is usually more complicated.
Some AI companies will likely become enormously valuable.
Others may disappear.
This is similar to the dot-com era.
The Internet succeeded.
Many Internet companies did not.

2. SoftBank

Masayoshi Son has extraordinary conviction.
Supporters say:

He sees the future earlier than everyone else.

Critics ask:

Is he confusing vision with certainty?

The documentary poses exactly this question.
If AI succeeds on the scale Son expects,
then today' s leverage may appear justified.
If AI disappoints,
the same leverage could amplify losses.
The difference between visionary investing and excessive optimism often becomes clear only in hindsight.

3. China

The same theme applies to China.

Excessive pessimism

Some investors believe:

China' s economy will never recover.
Property will never stabilize.
Consumption will remain weak indefinitely.

Excessive optimism

Others believe:

Government stimulus will solve every problem quickly.
Growth will return to previous highs.
Property prices will soon recover.

The reality is likely somewhere between those extremes.
Long-term investors benefit from distinguishing durable strengths from cyclical challenges rather than relying on either extreme narrative.

4. Currency markets

Your recent questions about the yen also fit this theme.
Markets can believe:

" The dollar will strengthen forever."

" The yen must rebound immediately."

Neither assumption is guaranteed.
Currencies are influenced by:

interest-rate differentials,
inflation,
productivity,
capital flows,
fiscal policy,
central-bank actions.

Intervention can influence short-term movements, but longer-term trends usually depend on underlying economic fundamentals.

5. Global liquidity

One common form of wishful thinking during a bull market is:

" Liquidity will always be available."

History suggests otherwise.
Examples include:

1998 (Long-Term Capital Management)
2008 (Global Financial Crisis)
2020 (COVID-19 market shock)
2023 (regional U.S. banking stress)

During periods of stress:

lenders become more cautious,
leverage becomes more expensive,
refinancing becomes harder,
markets place greater value on liquidity.

Applying the theme to your portfolio

Given the investments you' ve discussed previously, the lesson is not to avoid optimism but to avoid relying on optimism alone.
For example:

OCBC, DBS, and Hong Leong Finance are generally valued based on earnings, dividends, capital strength, and credit quality.
HSBC depends on global banking conditions and cross-border activity.
Ping An depends on insurance operations, investment returns, and China' s economic environment.
SoftBank depends more heavily on AI execution, asset valuations, and access to financing.

The key question for each investment is not simply " Will things improve?" but " What evidence supports that expectation, and what risks could prevent it?"

Strategic lesson

The theme of Wishful Thinking can be summarized as a timeless investing principle:

Hope is valuable because it encourages long-term investment. Wishful thinking becomes dangerous when it replaces disciplined analysis.

The most successful long-term investors&mdash including Warren Buffett, Li Ka-shing, and many enduring family holding companies you' ve been studying&mdash generally combine optimism about the future with careful attention to valuation, balance-sheet strength, liquidity, and downside risk. They prepare for favorable outcomes while also recognizing that markets rarely move in a straight line.

https://www.youtube.com/watch?v=ldQpRMegYc0& list=RDldQpRMegYc0& start_radio=1

Useful To Me Not Useful To Me

chartiskao
Supreme

04-Aug-2026 20:27

chartiskao About Contact Quote this Post!

x 0
x 0

Alert Admin

https://www.youtube.com/watch?v=yoy-o-lccKE

Here' s a concise strategic summary of the history of Google based on your notes.

The Evolution of Google: From University Research to Global Technology Leader

1. The Early Days of Search (1990s)

Before Google, internet search was inefficient.

Search engines such as AltaVista, Lycos, and Excite mainly ranked websites by counting how often keywords appeared.
This encouraged website owners to " stuff" pages with repeated keywords to manipulate rankings.
Search results were often inaccurate, irrelevant, and difficult for users.

At the same time, Yahoo dominated the internet through its portal strategy.
Yahoo believed users should:

Stay on Yahoo' s homepage.
Read Yahoo News, Finance, Sports, and Email.
Generate advertising revenue by spending more time within Yahoo' s ecosystem.

Searching was only one feature among many.

2. The Birth of Google (1996&ndash 1998)

Every website link acted as a " vote."
Votes from trusted websites carried greater weight.
Pages receiving many high-quality links ranked higher.

This approach measured authority rather than keyword repetition.
The result was dramatically better search quality.

3. The Missed Opportunity

Why?

Their business philosophies were fundamentally different.

Yahoo' s philosophy

Keep users inside Yahoo.
Increase page views.
Display more advertisements.

Google' s philosophy

Help users find exactly what they need as quickly as possible.
Let users leave Google immediately after getting useful results.

At the time, many executives believed Google' s approach would reduce advertising opportunities because users spent less time on the website.
Ironically, this became Google' s greatest strength.

4. Google' s Early Growth

5. The Revolutionary Revenue Model

Advertisers only paid when users clicked.
Ads were highly relevant to search queries.
Businesses could accurately measure return on investment.

This created a win-win situation:

Users received relevant advertisements.
Businesses reached interested customers.
Google earned revenue only when advertising created value.

The model proved enormously scalable and became one of the most successful advertising systems in history.

6. Going Public (2004)

Google' s Initial Public Offering (IPO) in 2004 became one of the most important technology listings ever.
Its successful public debut:

Generated billions of dollars.
Demonstrated that internet companies could build profitable businesses.
Helped restore investor confidence after the dot-com crash.

The IPO also provided Google with resources for long-term expansion.

7. Beyond Search

Google evolved from a search engine into a diversified technology company.
Major milestones included:

Acquiring YouTube in 2006, becoming the world' s largest video platform.
Developing the Android operating system, now powering billions of smartphones.
Expanding into cloud computing, artificial intelligence, digital maps, productivity software, autonomous driving research, and hardware.

In 2015, Google reorganized under its parent company, Alphabet Inc., allowing its various businesses to operate more independently.

8. Strategic Lessons

Google' s rise illustrates several important business principles:

Lesson	Explanation
Solve the user' s problem first	Superior user experience built long-term loyalty.
Better technology creates competitive advantage	PageRank produced significantly more relevant search results.
A strong business model amplifies innovation	AdWords transformed excellent technology into a highly profitable enterprise.
Short-term thinking can miss transformative opportunities	Yahoo and Excite underestimated Google' s disruptive potential.
Focus can outperform feature overload	Google concentrated on delivering the best search experience rather than becoming a broad internet portal.

Conclusion

chartiskao ( Date: 04-Aug-2026 14:25) Posted:

I can discuss the themes of the song and apply them as a metaphor to today' s AI bubble, China, SoftBank, and the global financial system.

" Wishful Thinking" as an investment philosophy

The title itself captures one of the biggest risks in financial markets:

Confusing hope with probability.

Every major bubble contains an element of wishful thinking.

1. AI Bubble

Wishful thinking

Investors assume:

AI demand will grow forever.
Every AI company will become profitable.
Capital will remain cheap.
Governments will always support AI.

2. SoftBank

Masayoshi Son has extraordinary conviction.
Supporters say:

He sees the future earlier than everyone else.

Critics ask:

Is he confusing vision with certainty?

3. China

The same theme applies to China.

Excessive pessimism

Some investors believe:

China' s economy will never recover.
Property will never stabilize.
Consumption will remain weak indefinitely.

Excessive optimism

Others believe:

Government stimulus will solve every problem quickly.
Growth will return to previous highs.
Property prices will soon recover.

The reality is likely somewhere between those extremes.
Long-term investors benefit from distinguishing durable strengths from cyclical challenges rather than relying on either extreme narrative.

4. Currency markets

Your recent questions about the yen also fit this theme.
Markets can believe:

" The dollar will strengthen forever."

" The yen must rebound immediately."

Neither assumption is guaranteed.
Currencies are influenced by:

interest-rate differentials,
inflation,
productivity,
capital flows,
fiscal policy,
central-bank actions.

Intervention can influence short-term movements, but longer-term trends usually depend on underlying economic fundamentals.

5. Global liquidity

One common form of wishful thinking during a bull market is:

" Liquidity will always be available."

History suggests otherwise.
Examples include:

1998 (Long-Term Capital Management)
2008 (Global Financial Crisis)
2020 (COVID-19 market shock)
2023 (regional U.S. banking stress)

During periods of stress:

lenders become more cautious,
leverage becomes more expensive,
refinancing becomes harder,
markets place greater value on liquidity.

Applying the theme to your portfolio

Given the investments you' ve discussed previously, the lesson is not to avoid optimism but to avoid relying on optimism alone.
For example:

OCBC, DBS, and Hong Leong Finance are generally valued based on earnings, dividends, capital strength, and credit quality.
HSBC depends on global banking conditions and cross-border activity.
Ping An depends on insurance operations, investment returns, and China' s economic environment.
SoftBank depends more heavily on AI execution, asset valuations, and access to financing.

The key question for each investment is not simply " Will things improve?" but " What evidence supports that expectation, and what risks could prevent it?"

Strategic lesson

The theme of Wishful Thinking can be summarized as a timeless investing principle:

Hope is valuable because it encourages long-term investment. Wishful thinking becomes dangerous when it replaces disciplined analysis.

https://www.youtube.com/watch?v=ldQpRMegYc0& list=RDldQpRMegYc0& start_radio=1

chartiskao ( Date: 03-Aug-2026 06:20) Posted:

Given what I know about your investment journey, your philosophy has been remarkably consistent. You began with Benjamin Graham' s value investing in 1975, then evolved toward high-quality dividend companies in Singapore, Hong Kong and China, and in recent years you' ve emphasized buying financially strong companies during crises and living off a growing stream of dividends rather than chasing fashionable sectors.
That background is important because the MAS article is not saying AI will collapse. It is saying that Singapore' s economy has become more exposed to one investment theme than many people realise.
Here' s how I would interpret it in light of your investment philosophy.

What MAS is really warning about

Many readers focus on the stock market volatility.
I think MAS is warning about something deeper.
The chain looks like this:
AI optimism
&rarr hyperscalers invest hundreds of billions
&rarr semiconductor demand surges
&rarr Singapore exports boom
&rarr manufacturing grows
&rarr employment improves
&rarr corporate profits rise
If one link weakens:
Hyperscalers reduce capex
&rarr semiconductor orders slow
&rarr Singapore electronics exports fall
&rarr manufacturing weakens
&rarr the broader economy slows.
MAS is essentially warning against assuming that today' s AI-driven growth will continue indefinitely.

This is not the first technology boom

During your investing lifetime, you have already seen several cycles.

Period	Popular theme	What happened
1970s	Nifty Fifty	Overvaluation followed by a long correction
Late 1980s	Japanese asset bubble	Collapse
Late 1990s	Internet boom	Dot-com crash
Mid-2000s	US housing	Global Financial Crisis
2020&ndash 2021	Pandemic technology	Sharp correction in 2022
2024&ndash 2026	AI infrastructure	Still unfolding

Every generation believes:

" This time is different."

Usually the technology changes the world.
But stock prices often move far ahead of business fundamentals.

Why your strategy has worked

Looking at your history, you rarely bought into market euphoria.
Instead you bought when:

banks were hated
REITs were sold aggressively
Hong Kong property developers traded below book value
dividend yields became unusually attractive

That is classic Graham thinking.
You bought fear.
Not excitement.

AI will probably change the world

I actually agree with the optimistic view.
AI may become as important as:

electricity
the internet
smartphones

But that does not mean every AI-related stock is a good investment at every price.
History shows:
Railroads changed America.
Many railroad shareholders still lost money.
The internet changed the world.
Many dot-com companies disappeared.
Technology success and investment success are not the same thing.

What concerns me more

The article mentioned something very important:
Microsoft
Amazon
Meta
Alphabet
Oracle
SpaceX
have committed enormous capital expenditure.
Those investments must eventually earn attractive returns.
Eventually investors will ask:
" Where is the profit?"
That is a healthy question.

How should you respond?

Given your investment philosophy, I would not make a dramatic change.
Instead I would continue applying the principles that have guided you successfully.

1. Continue holding quality dividend compounders

The Singapore banks still possess characteristics you have long valued:

strong capital
recurring earnings
disciplined management
sustainable dividends

Even if interest rates fall somewhat, they remain high-quality franchises.

2. Continue holding cash reserves

One of your greatest strengths has been maintaining liquidity for market downturns.
If AI enthusiasm turns into an overcorrection,
cash becomes valuable.

3. Avoid chasing AI winners after large rallies

When everyone believes:
" This company cannot lose,"
valuation deserves extra scrutiny.

4. Continue buying assets rather than stories

Companies with:

tangible assets
recurring cash flow
conservative balance sheets
growing dividends

often prove more resilient through cycles than businesses valued mainly on future expectations.

5. Think globally

Singapore banks
HSBC
Hong Kong property
Chinese insurers
all respond to different economic drivers.
Diversification across sectors and regions can reduce reliance on any single theme, including AI.

One thing I would reconsider

There is one area where I might gently refine your approach.
Historically, much of your portfolio has been concentrated in financials, property-related companies and dividend stocks. Those have served you well, especially when purchased at attractive valuations.
However, over the next 10&ndash 20 years, global productivity gains are likely to come increasingly from technology, healthcare and digital infrastructure. You do not need to chase expensive AI stocks, but you may benefit from having some exposure to broad technology or global innovation through diversified funds rather than relying solely on traditional sectors. This allows you to participate in long-term structural growth without taking concentrated valuation risk.

My conclusion

The MAS article is, in essence, saying:

Don' t mistake a cyclical AI investment boom for a permanent guarantee of economic growth.

That message aligns closely with the philosophy that has guided your investing for decades.
Benjamin Graham taught that the market alternates between greed and fear. Warren Buffett added that investors should be fearful when others are greedy and greedy when others are fearful.
Your investment history reflects those principles. Rather than abandoning them because of the AI boom, the current environment reinforces their relevance. Keep focusing on business quality, balance-sheet strength, sustainable dividends, reasonable valuations, and maintaining enough liquidity to take advantage of opportunities if today' s enthusiasm eventually gives way to tomorrow' s fear.

Useful To Me Not Useful To Me

chartiskao
Supreme

04-Aug-2026 14:25

chartiskao About Contact Quote this Post!

x 0
x 0

Alert Admin

I can discuss the themes of the song and apply them as a metaphor to today' s AI bubble, China, SoftBank, and the global financial system.

" Wishful Thinking" as an investment philosophy

The title itself captures one of the biggest risks in financial markets:

Confusing hope with probability.

Every major bubble contains an element of wishful thinking.

1. AI Bubble

Wishful thinking

Investors assume:

AI demand will grow forever.
Every AI company will become profitable.
Capital will remain cheap.
Governments will always support AI.

2. SoftBank

Masayoshi Son has extraordinary conviction.
Supporters say:

He sees the future earlier than everyone else.

Critics ask:

Is he confusing vision with certainty?

3. China

The same theme applies to China.

Excessive pessimism

Some investors believe:

China' s economy will never recover.
Property will never stabilize.
Consumption will remain weak indefinitely.

Excessive optimism

Others believe:

Government stimulus will solve every problem quickly.
Growth will return to previous highs.
Property prices will soon recover.

The reality is likely somewhere between those extremes.
Long-term investors benefit from distinguishing durable strengths from cyclical challenges rather than relying on either extreme narrative.

4. Currency markets

Your recent questions about the yen also fit this theme.
Markets can believe:

" The dollar will strengthen forever."

" The yen must rebound immediately."

Neither assumption is guaranteed.
Currencies are influenced by:

interest-rate differentials,
inflation,
productivity,
capital flows,
fiscal policy,
central-bank actions.

Intervention can influence short-term movements, but longer-term trends usually depend on underlying economic fundamentals.

5. Global liquidity

One common form of wishful thinking during a bull market is:

" Liquidity will always be available."

History suggests otherwise.
Examples include:

1998 (Long-Term Capital Management)
2008 (Global Financial Crisis)
2020 (COVID-19 market shock)
2023 (regional U.S. banking stress)

During periods of stress:

lenders become more cautious,
leverage becomes more expensive,
refinancing becomes harder,
markets place greater value on liquidity.

Applying the theme to your portfolio

Given the investments you' ve discussed previously, the lesson is not to avoid optimism but to avoid relying on optimism alone.
For example:

OCBC, DBS, and Hong Leong Finance are generally valued based on earnings, dividends, capital strength, and credit quality.
HSBC depends on global banking conditions and cross-border activity.
Ping An depends on insurance operations, investment returns, and China' s economic environment.
SoftBank depends more heavily on AI execution, asset valuations, and access to financing.

The key question for each investment is not simply " Will things improve?" but " What evidence supports that expectation, and what risks could prevent it?"

Strategic lesson

The theme of Wishful Thinking can be summarized as a timeless investing principle:

Hope is valuable because it encourages long-term investment. Wishful thinking becomes dangerous when it replaces disciplined analysis.

https://www.youtube.com/watch?v=ldQpRMegYc0& list=RDldQpRMegYc0& start_radio=1

chartiskao ( Date: 03-Aug-2026 06:20) Posted:

What MAS is really warning about

This is not the first technology boom

During your investing lifetime, you have already seen several cycles.

Period	Popular theme	What happened
1970s	Nifty Fifty	Overvaluation followed by a long correction
Late 1980s	Japanese asset bubble	Collapse
Late 1990s	Internet boom	Dot-com crash
Mid-2000s	US housing	Global Financial Crisis
2020&ndash 2021	Pandemic technology	Sharp correction in 2022
2024&ndash 2026	AI infrastructure	Still unfolding

Every generation believes:

" This time is different."

Usually the technology changes the world.
But stock prices often move far ahead of business fundamentals.

Why your strategy has worked

Looking at your history, you rarely bought into market euphoria.
Instead you bought when:

banks were hated
REITs were sold aggressively
Hong Kong property developers traded below book value
dividend yields became unusually attractive

That is classic Graham thinking.
You bought fear.
Not excitement.

AI will probably change the world

I actually agree with the optimistic view.
AI may become as important as:

electricity
the internet
smartphones

What concerns me more

How should you respond?

Given your investment philosophy, I would not make a dramatic change.
Instead I would continue applying the principles that have guided you successfully.

1. Continue holding quality dividend compounders

The Singapore banks still possess characteristics you have long valued:

strong capital
recurring earnings
disciplined management
sustainable dividends

Even if interest rates fall somewhat, they remain high-quality franchises.

2. Continue holding cash reserves

One of your greatest strengths has been maintaining liquidity for market downturns.
If AI enthusiasm turns into an overcorrection,
cash becomes valuable.

3. Avoid chasing AI winners after large rallies

When everyone believes:
" This company cannot lose,"
valuation deserves extra scrutiny.

4. Continue buying assets rather than stories

Companies with:

tangible assets
recurring cash flow
conservative balance sheets
growing dividends

often prove more resilient through cycles than businesses valued mainly on future expectations.

5. Think globally

Singapore banks
HSBC
Hong Kong property
Chinese insurers
all respond to different economic drivers.
Diversification across sectors and regions can reduce reliance on any single theme, including AI.

One thing I would reconsider

My conclusion

The MAS article is, in essence, saying:

Don' t mistake a cyclical AI investment boom for a permanent guarantee of economic growth.

chartiskao ( Date: 02-Aug-2026 00:41) Posted:

Strategic Valuation Report: Tencent (HKEX: 0700) vs US Mega-Cap Technology Companies (August 2026)

Executive Summary

Based on the valuation metrics you provided, Tencent is the cheapest large-cap technology platform among the world' s AI leaders despite possessing one of the strongest business ecosystems.

Company	P/E	Dividend Yield	Market Cap
Tencent	16.11x	1.12%	HK$4.32T
Alphabet	17.92x	0.25%	US$4.35T
Meta	20.96x	0.38%	US$1.41T
Microsoft	27.68x	0.78%	US$3.45T
NVIDIA	30.74x	0.50%	US$4.86T
Apple	37.37x	0.35%	US$4.54T
AMD	156.24x	0%	US$776B

1. Valuation Comparison

Tencent vs Meta

Metric	Tencent	Meta
P/E	16.11x	20.96x
Dividend	1.12%	0.38%
Relative valuation	23% cheaper	Premium

Assessment

Meta generates outstanding advertising profits and has improved efficiency significantly. However:

Meta depends heavily on advertising.
Tencent has multiple engines of earnings:
- Gaming
- WeChat ecosystem
- Payments
- Cloud
- Advertising
- Enterprise software
- AI
- Investment portfolio

Tencent therefore has broader diversification while trading at a lower multiple.
Winner: Tencent on valuation.

2. Tencent vs Alphabet

Metric	Tencent	Alphabet
P/E	16.11x	17.92x
Dividend	1.12%	0.25%

Alphabet is only slightly more expensive.
Both companies have enormous AI opportunities.
Alphabet dominates:

Search
Android
YouTube
Gemini AI

Tencent dominates:

WeChat
Gaming
China cloud
Payments
Mini-program ecosystem

Alphabet has stronger global exposure.
Tencent offers higher income and lower valuation.

3. Tencent vs Microsoft

Metric	Tencent	Microsoft
P/E	16.11x	27.68x

Microsoft trades at almost 72% higher valuation.
Why?
Microsoft enjoys:

Azure
Office
Windows
Copilot AI
Enterprise software

Investors pay a premium for recurring subscription revenue.
Tencent' s earnings are also recurring but remain discounted because of:

China regulatory concerns
Geopolitical risk
Lower foreign ownership

4. Tencent vs Apple

Metric	Tencent	Apple
P/E	16.11x	37.37x

Apple trades at more than double Tencent' s valuation.
Apple enjoys:

Premium hardware
Services ecosystem
Massive buybacks

However:

iPhone growth is slowing.
China smartphone competition is intensifying.
AI monetization remains uncertain.

Tencent currently appears much cheaper on earnings.

5. Tencent vs NVIDIA

Metric	Tencent	NVIDIA
P/E	16.11x	30.74x

NVIDIA deserves a premium because it dominates AI GPUs.
However:

Semiconductor cycles can be volatile.
Hardware margins may normalize over time.

Tencent is an AI application platform rather than an AI infrastructure provider, leading to different risk and growth profiles.

6. Tencent vs AMD

AMD trades at 156x earnings, reflecting expectations of substantial future growth.
Investors are pricing:

AI chips
Data-center expansion
Market share gains

Tencent is already highly profitable and generates significant free cash flow, while AMD' s valuation assumes strong future execution.

Growth Comparison

Company	Main Growth Drivers	Growth Outlook
Tencent	AI, WeChat ecosystem, gaming, fintech, cloud, advertising	High
Alphabet	Gemini AI, Search, YouTube, Cloud	High
Meta	AI advertising, WhatsApp monetization, Reality Labs	High
Microsoft	Azure, Copilot, Office AI	High
NVIDIA	AI chips, networking, enterprise AI	Very High
Apple	Services, AI-enabled devices, ecosystem	Moderate
AMD	AI accelerators, CPUs, GPUs	High but more cyclical

Dividend Comparison

Company	Dividend Yield
Tencent	1.12%
Microsoft	0.78%
NVIDIA	0.50%
Meta	0.38%
Apple	0.35%
Alphabet	0.25%
AMD	0%

Tencent provides the highest stated dividend yield among this group.

Business Quality Comparison

Company	Competitive Advantage
Tencent	WeChat ecosystem, gaming leadership, fintech, cloud, digital content
Microsoft	Enterprise software ecosystem, Azure, Office
Alphabet	Search, YouTube, Android, AI
Meta	Social networks, advertising, messaging
Apple	Hardware-software ecosystem, brand
NVIDIA	AI hardware leadership
AMD	High-performance CPUs and GPUs

Tencent' s ecosystem spans communication, payments, gaming, content, cloud, and AI, creating multiple revenue streams.

Risk Assessment

Tencent

China regulatory policy
US-China geopolitical tensions
Slower domestic economy

Microsoft

Rich valuation
Enterprise spending cycles

Alphabet

Antitrust scrutiny
AI competition

Apple

Hardware replacement cycles
China market exposure

NVIDIA

Semiconductor cycle
Customer concentration
High expectations embedded in valuation

AMD

Execution risk
Intense competition
Elevated valuation

Relative Valuation Ranking (Lowest P/E to Highest)

Rank	Company	P/E
1	Tencent	16.11x
2	Alphabet	17.92x
3	Meta	20.96x
4	Microsoft	27.68x
5	NVIDIA	30.74x
6	Apple	37.37x
7	AMD	156.24x

Dividend Yield Ranking

Rank	Company	Yield
1	Tencent	1.12%
2	Microsoft	0.78%
3	NVIDIA	0.50%
4	Meta	0.38%
5	Apple	0.35%
6	Alphabet	0.25%
7	AMD	0%

Strategic Conclusion

From the metrics provided:

Value: Tencent stands out with the lowest P/E (16.11x), suggesting the market assigns a lower valuation multiple than to most comparable global technology leaders.
Income: Tencent also offers the highest dividend yield in this comparison (1.12%), although US peers such as Meta and Alphabet have tended to emphasize share repurchases over dividends.
Growth: NVIDIA currently has the strongest AI infrastructure growth profile, while Microsoft, Alphabet, Meta, and Tencent each have substantial AI opportunities through software, cloud, advertising, or consumer platforms.
Risk: Tencent' s lower valuation likely reflects additional perceived risks, including Chinese regulatory uncertainty and geopolitical factors. US peers generally trade at higher multiples partly because investors view their regulatory and market environments differently.

For investors who prioritize a combination of reasonable valuation, diversified technology businesses, and cash income, Tencent appears attractively valued relative to many US mega-cap peers. However, that discount should be weighed against the higher country and policy risks associated with investing in Chinese technology companies. A sustained improvement in China' s economic outlook or investor sentiment could narrow Tencent' s valuation gap, but there is no guarantee this will occur.

Useful To Me Not Useful To Me

First < Newer 41-60 of 137 Older> Last

ShareJunction Version: 27 Nov 2020 ver- All Rights Reserved. Copyright ShareJunction Pte. Ltd. Disclaimer: All prices from are delayed. ShareJunction does not provide you with any financial advice. We are not into the business of providing any investment advice. See our Terms and Conditions and Privacy Policy of using this website. Data is delayed for varying periods of time depending on the exchange, but for at least 15 minutes. Copyright © SIX Financial Information Ltd. and its licensors. All Rights reserved. Further distribution and use by third parties prohibited. SIX Financial Information and its licensors make no warranty for information displayed and accept no liability for data and prices. SIX Financial Information reserves the right to adapt and/or alter this website at any time without prior notice.

Web design by FoundationFlux. Hosted with Signetique Cloud.