To The Moon

Since 2025, the U.S. AI rally has sent shockwaves through global capital markets. From computing chips to cloud infrastructure, from large language models to AI agents, the U.S. stock market’s “Magnificent Seven” ($Apple(AAPL)$ , $Microsoft(MSFT)$ , $Alphabet(GOOG)$ , $Amazon.com(AMZN)$ , $NVIDIA(NVDA)$ , $Meta Platforms, Inc.(META)$ , $Tesla Motors(TSLA)$ ) have become the center of attention.

At the same time, debate over whether an “AI bubble” has already formed has grown increasingly intense.

On one side is a trillion-dollar capital-expenditure “arms race”; on the other is the reality that many application-level business models have yet to achieve full commercial closure. Investors are asking whether the market is replaying the internet bubble of 2000.

Drawing on the views of multiple authoritative institutions and brokerages, this article offers a relatively rational breakdown.

I. Macro Investment Intensity: Still Some Distance from “Runaway” Levels

Xia Haoyang of GF Fund highlighted a key indicator at a public strategy conference:

Current U.S. AI investment accounts for roughly 1% of GDP, whereas historical peaks during electrification and the IT boom of the 1990s reached 1.5%–2%.

In other words, while the scale of AI investment in this cycle is indeed large, it has not yet reached the peak intensity seen in previous technology super-cycles.

Moreover, this AI cycle has not been accompanied by an IPO frenzy. Whether measured by total fundraising volume or the number of newly listed companies, activity remains far below the levels seen during the dot-com bubble. If the year 2000 represented a nationwide speculative technology mania, today looks more like a structural expansion led by a handful of dominant firms.

II. The Magnificent Seven’s Cash Flows: Expanding with Profits, Not Betting the Future

The market’s biggest concern is whether these technology giants are mortgaging the future.

At present, the Magnificent Seven’s capital expenditures account for roughly 15% of revenue — admittedly on the high side. But the more critical metrics are:

• Capital Expenditure / Operating Cash Flow ≈ 50%

• Peak during the dot-com bubble ≈ 70%

At the same time, the ratio of free cash flow to long-term debt has generally remained above 70%. In contrast, near the end of the internet bubble this ratio once fell rapidly into single digits.

This implies that today’s expansion is primarily funded by internal profitability rather than high-leverage external financing. Brokerage research likewise notes that most AI capital spending by North American cloud providers is supported by their own cash flows, with no significant deterioration in cash-flow quality or leverage ratios. This stands in stark contrast to the many internet companies of the past that expanded without having achieved profitability.

III. Valuation Comparison: Far from the Frenzy of 2000

Another frequently discussed issue is valuation. The current price-to-earnings ratios of the top ten U.S. companies by market capitalization are roughly 30–40x, compared with about 20–30x for other firms.

By comparison, in 2000:

• Leading technology companies often traded at 80x earnings or higher

• The valuation gap versus traditional industries widened to extreme levels

More importantly, the recent rally has been driven more by earnings growth than by pure multiple expansion. Over the past two years, share-price increases among market leaders have been highly correlated with EPS growth. If the internet bubble was characterized by “valuations pulling prices upward,” the current cycle is more accurately described as “earnings supporting prices.”

Of course, certain niche segments have already priced in several years of future growth, but this represents structural overvaluation, not a systemic bubble

IV. The Real Debate: Can the Application Layer Achieve a Commercial Closed Loop?

Many fund companies have expressed similar views at annual strategy meetings: do not casually label it a bubble, but remain vigilant about the pace of monetization. Yao Zhipeng of Harvest Fund pointed out that current AI investment is still concentrated in computing power and infrastructure, while a truly dominant AI application giant has yet to emerge.

Industrial Securities also argues that the segments with greater long-term industrial potential lie in downstream applications, including:

• Financial risk control

• Manufacturing optimization

• Healthcare scenarios

• Industrial automation

If the application layer can generate sustainable cash flows, the logic behind massive AI capital expenditures will form a closed loop.

If monetization proceeds more slowly than expected, the market will undergo periodic corrections. This is the market’s true point of divergence — not whether the technology itself has value.

V. Conclusion: More Like the Early Stage of Technological Expansion Than the End of a Bubble

Looking back at the dot-com bubble, several classic characteristics stand out:

• Extremely weak profitability

• Extreme valuation expansion

• Large numbers of companies dependent on financing for survival

• IPO mania

• Rapid deterioration of cash flows

The current AI industry has not exhibited these signals on a comprehensive basis. Capital spending is aggressive, but still covered by profitability; valuations carry premiums, but remain far below historical extremes; financing structures are relatively sound, with no systemic leverage risk.

Therefore, rather than calling this a bubble, a more accurate description may be: an early-stage technology investment cycle characterized by high volatility and strong expectation-driven pricing.

The AI sector will inevitably experience pullbacks and divergence, but from the perspective of financial structure and industrial progress, it does not yet constitute a “full-scale bubble.”

For community investors, what truly matters is not debating whether a bubble exists, but distinguishing:

• Which companies are expanding on the strength of real cash flows

• Which sub-sectors are already overcrowded

• Which application scenarios are likely to achieve genuine monetization over the next three years

This AI rally is not a test of imagination, but a test of one’s ability to judge the pace at which expectations turn into reality.

For SG users only, welcome to open a CBA today and enjoy access to a trading limit of up to SGD 20,000 with unlimited trading on SG, HK, and US stocks, as well as ETFs.

🎉Cash Boost Account Now Supports 35,000+ Stocks & ETFs – Greater Flexibility Now

Find out more here.

Complete your first Cash Boost Account trade with a trade amount of ≥ SGD1000* to get SGD 688 stock vouchers*! The trade can be executed using any payment type available under the Cash Boost Account: Cash, CPF, SRS, or CDP.

Click to access the activity

other helpful links:

• 💰Join the TB Contra Telegram Group to Get $10 Trading Vouchers Now🎉

• How to open a CBA. How to link your CDP account. Other FAQs on CBA. Cash Boost Account Website.