Not Another Dot-Com Bubble? AI Spending Fuels a Broader Market Recovery

Author’s Note and Disclaimer:

Zion Zhao Real Estate | 88844623 | 狮家社小赵 | wa.me/6588844623 |  https://linktr.ee/zionzhao

This post is for general information, education, and market literacy only. It does not constitute financial, investment, trading, legal, tax, accounting, or other professional advice, and is not an offer, solicitation, recommendation, or endorsement. Views expressed are personal, general in nature, and subject to change without notice. While reasonable care is taken, no representation or warranty is given as to accuracy, completeness, or reliability. Readers should conduct independent due diligence and seek professional advice. To the fullest extent permitted by law, no liability is accepted for any loss arising from reliance on this material. 

AI Capex Boom Reprices Markets as Investors Look Past Dot-Com Comparisons

What If the AI Bull Market Is Still Early?

The most important question in today’s market is not whether equities feel expensive. They do. The sharper question is whether earnings, margins, productivity and cash flows are improving fast enough to justify higher prices. Episode 244 of The Compound and Friends, featuring Fidelity’s Denise Chisholm, frames this debate with unusual clarity: the AI capex boom may look dramatic, but large numbers alone do not make a bubble.

The market’s bull case is no longer built purely on enthusiasm. Goldman Sachs’ S&P 500 target of 8,000 is best understood as an earnings forecast, not a sentiment forecast. Its thesis depends on stronger profit estimates, AI infrastructure demand and broader participation across corporate America (Goldman Sachs, 2026). That distinction matters. A market driven mainly by valuation expansion is fragile. A market where earnings rise faster than prices may be more disciplined than headlines suggest.

This is why valuation compression can be quietly bullish. If earnings estimates rise while share prices do not fully keep pace, forward valuation multiples decline. In simple terms, the market becomes cheaper relative to expected profits even as index levels remain high. That is not classic bubble behaviour. It is a market climbing a wall of worry, supported by scepticism rather than euphoria.

The AI capital expenditure cycle also deserves a more serious reading. Today’s largest spenders are not fragile dot-com start-ups funded by speculative equity issuance. They are profitable technology platforms with deep balance sheets, massive customer bases, dominant distribution channels and significant free cash flow. Their spending is not abstract. It is physical, strategic and industrial. Data centres, semiconductors, memory chips, cloud infrastructure, networking equipment, cooling systems, power supply and grid access are turning AI from a software story into an infrastructure story.

That is why Big Tech increasingly resembles a group of “modern industrials.” These companies still enjoy software-like scalability and platform economics, but they now require industrial-scale investment. The AI era is not just about algorithms. It is about energy, land, chips, compute, capital intensity and execution. This changes how investors should think about margins, returns on invested capital and long-term valuation.

Still, discipline is essential. Capex is only value-creating if it converts into revenue, productivity, margin resilience and durable cash flow. If AI demand disappoints, the same spending that looks visionary today could be repriced as overcapacity tomorrow. The correct question is not whether AI is important. It clearly is. The correct question is whether the spending earns its cost of capital.

Semiconductors sit at the centre of this tension. The sector has seen extraordinary price action, especially in memory and AI-related chips. Bulls argue that demand is real, supply is tight and earnings growth is validating the move. Bears warn that semiconductors remain cyclical and that investors may be extrapolating peak conditions too far into the future. Both sides have merit. AI demand is genuine, but extreme price moves reduce the margin for error.

The most underappreciated market story, however, may be outside the Magnificent Seven. FactSet data show that earnings growth is broadening beyond mega-cap technology, weakening the argument that this market is supported by only a handful of stocks (FactSet, 2026). If the S&P 493, meaning the rest of the index excluding the largest technology leaders, is only beginning to recover after a multi-year earnings slowdown, the broader market cycle may be earlier than many investors assume.

This matters because narrow leadership is fragile, but broadening earnings are healthier. If smaller, cyclical, financial, industrial, consumer and non-tech companies begin contributing more meaningfully to profit growth, the market becomes less dependent on a few mega-cap winners. That is a powerful shift in market structure and sentiment.

Inflation and Federal Reserve policy remain important risks, but they are not automatic market killers. Inflation is still above the Federal Reserve’s target, and policy uncertainty remains relevant (U.S. Bureau of Labor Statistics, 2026; Federal Reserve, 2026). Yet equities can still perform in a moderate inflation environment if nominal earnings growth, productivity and margins remain strong. The real danger is not inflation alone. The real danger is inflation that becomes broad-based, persistent and severe enough to force aggressive tightening while earnings weaken.

The coming mega-IPO wave may become the next major test. Companies such as Anthropic, OpenAI and SpaceX could force public markets to absorb enormous new supply. That may create short-term liquidity pressure as funds rebalance and raise cash. But it may also provide greater transparency. Public markets will demand audited financials, disclosure, margin data, customer concentration analysis and clearer profitability pathways. The AI story will have to move from private-market narrative to public-market evidence.

The conclusion is balanced. This is not risk-free. Semiconductor momentum may be stretched. AI capex may become excessive. Inflation may remain sticky. Mega-IPOs may challenge liquidity. Valuations may leave little room for disappointment.

But this is not obviously 2000 either. Many of today’s AI leaders are profitable, cash-generative and strategically entrenched. Earnings are broadening. Capex is tied to real demand. The market remains sceptical, not euphoric.

The AI era may still be early. But early does not mean easy. It means serious investors must separate durable earnings power from narrative excess, productive infrastructure from speculative overbuild, and genuine broadening from temporary momentum.

References

FactSet. (2026). “Mag 7” and other 493 S&P 500 companies are reporting highest earnings growth since 2021.

Federal Reserve. (2026). Minutes of the Federal Open Market Committee, April 28 to 29, 2026.

Goldman Sachs. (2026). The S&P 500 is forecast to climb as earnings growth powers stocks higher.

The Compound and Friends. (2026). What if it’s still early? With Denise Chisholm.

U.S. Bureau of Labor Statistics. (2026). Consumer Price Index, April 2026.

What If the AI Bull Market Is Still Early? Why Earnings, Infrastructure and Real Assets Matter Now

The AI bull market may not be ending. It may be broadening. The stronger case is no longer hype, but earnings resilience, AI infrastructure demand, margin discipline and recovery beyond the Magnificent Seven. Risks remain in semiconductors, inflation and mega IPO liquidity, but this looks less like 2000 and more like an early industrial productivity cycle.

For Singapore property buyers, sellers, landlords, tenants and investors, the AI market debate is not just about stocks. It is about how capital, productivity, infrastructure and confidence shape asset values.

If AI capex becomes a durable industrial cycle, demand may extend beyond technology shares into data centres, business parks, logistics hubs, power infrastructure, advanced manufacturing, office transformation and high-quality residential demand near employment nodes. For Singapore, this matters. Our property market is deeply connected to global liquidity, corporate expansion, interest rates, talent flows and long-term confidence in Asia as a capital hub.

For buyers, the lesson is timing with discipline. Do not chase hype, but do not ignore structural growth corridors.

For sellers, stronger capital markets and improving business sentiment can support wealth creation, upgrading demand and investor appetite.

For landlords, AI-driven business formation and corporate expansion may reshape tenant demand across commercial, industrial and residential sectors.

For investors, the key is selectivity. The best opportunities are not always the loudest. They are often found where infrastructure, policy, connectivity, scarcity and future earnings power meet.

This is why property decisions should not be made from headlines alone. They should be guided by macroeconomics, capital flows, land scarcity, rental fundamentals, policy risk and long-term asset positioning.

As a Singapore real estate salesperson with a strong grounding in markets, economics, asset allocation and property strategy, I help clients connect global trends with practical property decisions.

If you are buying, selling, renting or investing in Singapore property, speak with me for a clear, objective and strategy-driven discussion.

Like, collect and subscribe to my social media channels for more insights on Singapore property, macro trends and wealth positioning.

This is educational commentary only and not financial or investment advice.