The AI Trade Is Evolving: Why the Next Winners Will Be Built Beyond Chips

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.  

The AI Trade Is Not Cooked. It Is Moving Up the Stack.

Artificial intelligence investing is entering a more demanding phase.

For the past three years, investors debated whether generative AI was a genuine technological revolution, whether enterprises would spend aggressively, and whether semiconductor demand could justify extraordinary valuations.

Those questions are increasingly being answered.

The harder question now is: who captures the economic value after the infrastructure is built?

NVIDIA’s fiscal second-quarter 2027 results illustrate the scale of the AI infrastructure boom. Revenue reached US$96.2 billion, while GAAP net income approached US$59.7 billion. Demand visibility remains exceptionally strong, but that does not mean semiconductor analysis is finished. Investors must still consider valuation, margins, customer concentration, custom accelerators, export controls, supply-chain concentration, power availability and capital-cycle risk (NVIDIA, 2026; U.S. Government Accountability Office [GAO], 2026).

Amazon reinforces the scale of the buildout. AWS plans to deploy two million additional NVIDIA GPUs during 2027 and 2028. Yet Amazon is simultaneously investing in custom silicon, databases, robotics and Zoox autonomous vehicles. The message is clear: AI value creation is spreading beyond chips into cloud infrastructure, proprietary data and physical automation.

Apple represents another emerging frontier. Its latest Mac hardware is increasingly designed for on-device and agentic AI. Edge computing offers advantages in privacy, latency, offline processing and local data control, although cloud infrastructure remains essential for the largest models and most demanding workloads (Apple, 2026a, 2026b; Wang et al., 2025). The future is unlikely to be cloud versus edge. It will be cloud plus edge.

Meta and Grok Bot highlight perhaps the most important transition of all: AI is moving from answering questions to executing work.

A chatbot responds. An AI agent acts.

Future agents may research products, manage calendars, build storefronts, reconcile invoices, communicate with businesses and execute complex digital workflows. If these systems become reliable, AI could begin competing not merely for software budgets, but for portions of global labour expenditure.

However, productivity is not automatic. Research has repeatedly shown that technology creates the greatest economic benefits when organisations redesign workflows, incentives and management around it (Brynjolfsson & Hitt, 2000; Brynjolfsson et al., 2023).

Physical AI is emerging simultaneously. Waymo, Zoox and Tesla are competing to commercialise autonomous transportation, while Amazon increasingly integrates robotics into logistics. Tesla and SpaceX are also pursuing ambitious semiconductor vertical integration, while SpaceX is exploring enormous infrastructure projects and potentially orbital computing.

These opportunities are real, but so are the engineering constraints. Space-based AI computing, for example, does not eliminate cooling requirements. In vacuum, waste heat must ultimately be rejected through thermal radiation, making thermal control a major engineering consideration (NASA, 2026).

The broader investment lesson is therefore straightforward.

The first phase of the AI boom was about supplying compute.

The next phase will increasingly be about monetising intelligence.

Investors should now look beyond GPU shipments and examine AI revenue, infrastructure utilisation, returns on invested capital, proprietary data, distribution, agent reliability, autonomous-vehicle economics and actual productivity gains.

The best AI model may not become the biggest AI business.

Technology history shows that enduring winners combine innovation with distribution, ecosystems, execution, switching costs and disciplined capital allocation.

The AI trade is not finished.

It is becoming more sophisticated.

And the ultimate winners will not simply be those building the most powerful models or largest data centres. They will be the companies that convert artificial intelligence into durable cash flow, sustainable competitive advantage and superior returns on invested capital.

References

Apple. (2026a). Apple unveils a more powerful Mac mini featuring the all-new M6 and M5 Pro. Apple Newsroom.

Apple. (2026b). Apple introduces new Mac Studio with M5 Max and M5 Ultra. Apple Newsroom.

Brynjolfsson, E., & Hitt, L. M. (2000). Beyond computation: Information technology, organizational transformation and business performance. Journal of Economic Perspectives, 14(4), 23–48.

Brynjolfsson, E., Li, D., & Raymond, L. R. (2023). Generative AI at work. National Bureau of Economic Research.

National Aeronautics and Space Administration. (2026). Thermal control. Small Spacecraft Systems Virtual Institute.

NVIDIA Corporation. (2026). NVIDIA announces financial results for second quarter fiscal 2027. NVIDIA Investor Relations.

U.S. Government Accountability Office. (2026). Semiconductors: Information on projects funded to strengthen U.S. supply chain.

Wang, X., Tang, Z., Guo, J., Meng, T., Wang, C., Wang, T., & Jia, W. (2025). Empowering edge intelligence: A comprehensive survey on on-device AI models. ACM Computing Surveys, 57(9).

AI Boom Shifts Beyond Chips as Investors Hunt the Next Profit Pools

Build Your Singapore Property Strategy With a Broader Investment Perspective

Singapore property should not be assessed in isolation. Interest rates, global capital flows, geopolitics, equity markets, currencies, regulation and economic cycles can all influence property values, financing costs, rental demand and portfolio outcomes.

As a Singapore real estate professional, I combine property advisory with experience in macroeconomics, global affairs, asset allocation, portfolio construction, equity and cryptocurrency markets, technical analysis, and Singapore land and business law. I also serve as an Officer Commanding in the Singapore Armed Forces at the rank of Captain, an appointment that reinforces discipline, responsibility and structured decision-making.

I dedicate hours daily to researching markets, conducting due diligence and writing analytical essays so that my clients can make decisions with broader context rather than relying solely on property headlines.

I welcome Singapore and international investors, China and Southeast Asian families, UHNW individuals, Family Offices exploring 13O, 13U, VCC or GIP-related considerations, institutional investors, parents accompanying students, and families planning relocation or long-term asset allocation.

Real estate can complement a diversified portfolio through tangible ownership, potential rental income and comparatively lower day-to-day price volatility than many listed assets, although returns and capital appreciation are never guaranteed.

If you value research-led, cross-asset and long-term property advice, I would be honoured to assist.