By Stephania Constantinou
Early this month, the S&P 500 rose 0.6 percent to a new high, taking its gain for the year to more than 14 percent. The main reason is the build-out and deployment of AI, which is lifting expectations of economic growth. Stocks have held firm despite the unpredictable war with Iran, a rise in US government bond yields, and news of AI agents going rogue. Yet, investors are piling into big tech companies building advanced frontier models such as AI.
The technology sector is expected to grow even more aggressively, and with it its influence on the index and the entire global economy. Over the last couple of years, the surge of AI models has made the market more concentrated and less diversified, with Nvidia, Alphabet and Amazon among its biggest holdings.
Yet concerns about AI models scheming and acting deceptively, along with mounting discussion among governments about safe and responsible AI, may also have contributed to the growing interest in the potential of these technologies and, thus, in investments. Contradictory at this may sound, it is a possibility. As someone interested in both political economy and AI’s impact on society, I was skeptical at first. Correlation is not causation. But as someone researching AI policy and governance, I found data that gave me pause.
Investment in AI infrastructure and hyperscalers is now the market’s crucial growth engine, making up 53 per cent of the S&P 500’s weighting. AI adoption has also become a main agenda item in boardrooms, even as corporate AI governance lags and boards lack the technical expertise to judge the risks of rapidly advancing frontier models. In 2025, 83 per cent of S&P 500 companies disclosed AI as a material risk, up from 12 per cent in 2023. The more frequent discussions are about trustworthy and reliable AI, the bigger the call for transparency and the larger the allocation of investor funds in companies they see follow suit. Data from PwC’s 2025 Global Investor Survey, show that investors want to invest in companies that leverage AI to scale across enterprise, while strengthening credible governance and risk controls. It is a matter of building confidence in corporate resilience, while jumping ‘S’ curves. According to The Conference Board, a business think tank, more than 60 per cent of executives believe that the development of responsible AI governance principles, disclosure and regulation are essential for wider adoption and investments.
Two trends are moving together. AI-linked stocks, from semiconductor and chip makers, to data centers and leading AI labs, dominate the market, due to the large capital expenditure tied to the huge supply chain dedicated to the AI buildout. Recent macroeconomic data do not point out to a slowdown, for the time being. At the same time, corporate leaders are increasingly worried about AI’s operational risks and want a regulatory framework for its safe and responsible use. Cybersecurity and reputational risks top the list of concerns for S&P 500 companies and investors, which strengthens the case for stronger guardrails and governance.
The call for responsible AI is also global, driven by concerned citizens everywhere, and ethical questions are central to frontier AI research, with many organizations and universities pushing for more ethical and humane systems. For companies, a culture of reliable AI is not only a wise business strategy but an incentive to innovate and test models rigorously before deployment. Even some AI company leaders have called for slowing development. As models become more regulated, market confidence may rise too, though that is hard to quantify as models continue to evolve rapidly.
Institutions are acting as well. The United Nations’ Global Dialogue on AI Governance convenes governments to discuss the risks of unregulated AI and advocate for bigger AI oversight through a formal global framework of safety standards, similar to those in aviation. In Singapore, where I am based, the International Scientific Exchange on AI Safety is putting AI safety research and governance on the public policy agenda, and the 2026 Singapore Consensus sets research priorities on AI risk assessment, development, control and societal resilience. In Europe, the EU AI Act, passed in 2024, is the world’s first comprehensive legal framework for AI, taking a risk-based approach to governance.
The parallel between a surprisingly strong S&P 500 and growing demand for AI transparency and security is hard to ignore. The global conversation about AI risk is a welcome by-product of the AI frenzy, spreading awareness in boardrooms and among the public of how the economy, technology, social welfare and governance are intertwined. For now, that may be showing up, directly or indirectly, in the market’s gains.