The New AI Capital Stack: Why SpaceX, Nvidia, Bond Yields and Public Trust Now Shape the Same Market

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 Is No Longer a Tech Story: It Is a Capital, Infrastructure and Trust Test

AI is no longer a narrow technology story. It has become the organising layer of a new capital regime, where elite research talent, frontier models, semiconductors, memory, energy, data centres, satellites, regulation, public trust and bond yields now trade as one interconnected system. The podcast’s deeper insight is that Andrej Karpathy joining Anthropic, Nvidia’s post-earnings volatility, SpaceX’s potential trillion dollar scale-up, America’s AI backlash, Trump’s paused AI oversight push and rising macro pressure are not isolated headlines. They are different signals from the same structural transformation.

Karpathy’s move to Anthropic matters because frontier AI is still constrained by rare research judgment, not merely GPU supply. Compute can be purchased, although at extraordinary cost. Data can be licensed, generated, captured or synthesised. But the ability to decide which experiments matter, which model behaviours are meaningful, which evaluations are misleading and which architecture paths can compound remains scarce. His role in pre-training and AI-assisted model improvement reinforces a broader point: the next phase of AI competition will not be won only by companies that spend the most. It will be won by companies that industrialise research, compress iteration cycles and convert compute into durable capability.

The public backlash against AI is equally important. Americans are not simply rejecting technology. They are questioning asymmetry. A small group of companies is raising massive capital, buying scarce power, recruiting elite researchers and creating trillion dollar valuations, while workers hear executives discuss automation, flatter organisations and AI-driven productivity. Pew Research Center data shows U.S. adults remain far more concerned than excited about AI’s growing role in daily life (Pew Research Center, 2026). The IMF has also warned that AI may affect a large share of jobs in advanced economies, with both augmentation and displacement risks (International Monetary Fund, 2024). This is why the industry’s legitimacy problem cannot be solved by better slogans alone. Productivity gains are real, but public acceptance depends on whether workers, consumers and communities see tangible upside.

The academic evidence is more balanced than the market narrative. Studies show generative AI can improve productivity, especially in writing, customer support and knowledge work, but the benefits depend heavily on workflow design, human oversight, training and incentives (Brynjolfsson et al., 2025; Noy & Zhang, 2023). AI can empower workers, compress learning curves and improve service quality. It can also displace roles, concentrate returns and amplify managerial pressure if deployed purely as a cost-cutting tool. The decisive question is not whether AI raises productivity. It is who captures the productivity dividend.

SpaceX’s two trillion dollar case is best understood as an infrastructure thesis. Starlink gives recurring connectivity revenue. Launch provides cost control and strategic logistics. AI data centres could turn SpaceX into a compute infrastructure platform. The reported Anthropic compute arrangement highlights how the hunger for large-scale AI capacity may reshape SpaceX’s financial narrative. But the valuation case requires discipline. Investors should separate current revenue, near-term execution and long-dated optionality. Starlink and launch are real businesses. Terrestrial AI compute may become a major growth engine. Orbital compute remains visionary, but it should still be treated as an option, not a guaranteed underwriting base case.

Nvidia remains the clearest barometer of the AI buildout. Its fiscal first quarter 2027 results were extraordinary, with revenue growth, data centre strength, large buybacks and a major dividend increase (Nvidia, 2026). Yet the stock reaction shows that even exceptional growth can struggle when expectations are already extreme. Markets do not reward greatness alone. They price future cash flows against valuation, competition, customer concentration, China exposure, margin sustainability and interest rates. Nvidia may still be a world-class company, but investors must recognise that the higher the expectation, the narrower the margin for disappointment.

The broader AI supply chain also looks increasingly cross-sectional. Chips, memory, networking, cooling, power equipment and data centres cannot all be valued on inconsistent assumptions forever. If the AI buildout is real and durable, some parts of the infrastructure stack may still be underappreciated. If demand slows, the most aggressively priced beneficiaries may underperform. The correct analytical lens is therefore not a simplistic “AI bubble” or “AI boom” label. It is return on invested capital across the full stack.

That is where macro becomes unavoidable. AI cannot escape the bond market. Higher oil prices, inflation and Treasury yields raise capital costs, increase energy sensitivity and compress long-duration valuations. A low-rate world can support distant cash flow narratives. A higher-rate world demands nearer-term monetisation, disciplined capital allocation and credible free cash flow. The more capital-intensive AI becomes, the more it must prove that demand can translate into economic returns.

The policy debate follows the same logic. The United States wants AI leadership, especially in competition with China. But speed without trust creates backlash, and trust without speed risks strategic decline. Sensible governance should avoid both extremes: suffocating regulation that protects incumbents and laissez-faire deployment that undermines public confidence. The better framework is layered: safety evaluations for the most powerful systems, sector-specific rules, data protection, liability clarity and international coordination on misuse risks.

My bottom line is clear. AI’s next phase will not be won merely by building better models. It will be won by those who control scarce infrastructure, prove productivity, manage policy risk, secure energy, maintain valuation discipline and earn public trust. Capability alone is not enough. The durable winners will combine capability, infrastructure, legitimacy and macro awareness.

References

Brynjolfsson, E., Li, D., & Raymond, L. R. (2025). Generative AI at work. Quarterly Journal of Economics.

International Monetary Fund. (2024). AI will transform the global economy. Let’s make sure it benefits humanity.

Noy, S., & Zhang, W. (2023). Experimental evidence on the productivity effects of generative artificial intelligence. Science.

Nvidia. (2026). NVIDIA announces financial results for first quarter fiscal 2027.

Pew Research Center. (2026). What the data says about Americans’ views of artificial intelligence.

Reuters. (2026). SpaceX accelerates IPO timeline, targets June listing on Nasdaq.

From Nvidia to SpaceX: The AI Boom Is Now a Test of Cash Flow, Compute and Confidence

AI is no longer a narrow technology story; it is the capital stack, linking talent, compute, chips, energy, satellites, regulation and trust. SpaceX, Nvidia and Anthropic show upside; public backlash, policy risk and higher yields define constraint. Winners will convert capability into durable cash flow, legitimacy and resilience.

AI, Nvidia, SpaceX, bond yields and global policy shifts may look far removed from Singapore property, but they are increasingly connected to how capital moves, how confidence forms and how buyers, sellers, landlords, tenants and investors make decisions.

When AI infrastructure demand drives energy, chips, data centres and global capital expenditure, it affects inflation expectations, interest rates, equity market sentiment and liquidity. These forces influence mortgage affordability, investment appetite, rental demand, corporate expansion and cross-border wealth allocation into stable markets like Singapore.

For buyers, this means timing, financing discipline and asset selection matter more than ever. For sellers, pricing strategy must reflect not only recent caveats, but also macro sentiment, buyer psychology and liquidity conditions. For landlords and tenants, technology-driven business transformation may reshape leasing demand, tenant profiles and rental resilience. For investors, Singapore property should be assessed within a broader asset allocation framework, not in isolation.

As a Singapore real estate agent with a strong foundation in macroeconomics, global affairs, asset allocation, portfolio strategy, market cycles and Singapore property law, I help clients interpret the bigger picture before making property decisions. My role is not just to transact. It is to help you evaluate risk, opportunity, timing, pricing and negotiation with sharper market intelligence.

Whether you are buying, selling, renting or investing in Singapore property, work with someone who understands both real estate fundamentals and the global forces shaping capital.

For objective, data-informed and strategy-driven Singapore property guidance, engage my services today.

Like, collect and subscribe to my social media channels for more professional insights on Singapore property, macro trends and investment strategy.

Disclaimer: This content is for general market education only and is not financial, legal or tax advice. Please seek qualified professional advice before making decisions.