Artificial intelligence (AI) has become a recurring topic of discussion with investment firms. When asked whether we have reached the peak of investment in this theme—from both fixed income and equity perspectives—they insist we have not. They argue that there is still room and investment opportunities left within the AI universe, even following warnings from key industry figures—such as Dario Amodei (CEO of Anthropic), Sam Altman (CEO of OpenAI), and Elon Musk (xAI)—regarding the need to slow down development due to safety risks.
The impact of the current wave of investment in artificial intelligence could exceed $20 trillion. According to Capital Group, tech megacap capital expenditure is accelerating at a rate that could eclipse China’s industrialization process, referencing a benchmark of reaching $30 trillion by 2032.
“There is no doubt that artificial intelligence is becoming one of the primary drivers of economic activity. However, it should not be understood solely as a technology theme, but as an investment cycle with broad implications for the economy as a whole. The artificial intelligence ecosystem spans multiple levels, from semiconductor design and software development to power supply, infrastructure construction, and sectors integrating the new technology into their operations, such as media and financial services,” they note.
Debate or Marketing?
This massive investment opportunity has coincided in recent days with suggestions to moderate the pace of technological development, which had a slight impact on semiconductor companies while favoring software firms. “Leading model developers have little incentive to voluntarily slow down a technology they view as strategic, especially when Chinese competitors are just six to eight months behind,” according to Banca March.
This debate, combined with a higher interest rate outlook, could, according to Banca March’s latest analysis, “become the perfect backdrop for a temporary pullback in equity markets.” However, the firm’s experts downplay the concern, noting that “the narrative surrounding a potential slowdown in AI development seems to reflect an institutional marketing strategy ahead of two of the largest IPOs in history rather than an operational reality.”
“Competition in this space is extraordinarily intense, global, and decentralized, making any coordination attempt among primary players extremely difficult. Even more so when the Trump administration has been openly opposed, ruling out government interventions in the sector,” they add.
In the view of Flavien del Pino, Head of BDL Capital Management for Spain, the recent correction in tech companies most exposed to AI is not merely a cyclical market movement, but reflects a fundamental doubt regarding the actual profitability of this technology.
“Hyperscaler spending is accelerating to unprecedented levels and is destroying free cash flow generation, accumulating debt that will approach $2 trillion. To justify the $7 trillion that will be invested in data centers through 2030 with a return on capital employed (ROCE) of just 10%, the sector would need to generate $3.6 trillion annually in new revenues—a figure higher than the entire current global market for software and IT services,” he explains regarding the resulting capital return uncertainty.
Brakes on Investment
So, is there any factor that could genuinely stall AI investment? According to experts, a key issue will be national regulations—specifically, the outlook for AI regulation in the United States and potential restrictions on data center development. In the view of Libby Cantrill, Head of Public Policy at PIMCO, while the U.S. Congress may begin to focus more intensely on AI safety and federal government involvement appears inevitable at some point, “we are unlikely to see a comprehensive federal regulatory framework enacted into law in the near term.” Looking ahead to the next Congress, she notes there will likely be greater scrutiny on the issue, though “for now, AI regulation does not appear imminent.”
In the absence of federal progress, Cantrill believes “states are likely to continue moving forward with AI safety legislation” and taking the lead on data center restrictions. In this domain, municipalities in 32 states have already moved forward with moratoria, and up to 26 states are currently considering statewide moratoria.
Against this backdrop, Cantrill anticipates that “in 2027, given the political landscape, we could see greater friction in AI infrastructure development, with a likely widespread increase in data center construction costs” and, in some cases, states opting to halt them entirely. This would imply, she concludes, “a more complex patchworks for both companies and investors.”
Implications for Investors
From an investor’s perspective, Andrew Heiskell, Equity Strategist at Wellington Management, and Brian Barbetta, Global Industry Analyst at Wellington Management, consider that the debate is no longer centered on whether AI is relevant, but on a more complex question: which links in the ecosystem will capture the value generated?
“In such a dynamic environment, long-term technological progress and short-term public market expectations are unlikely to move at the same pace. Instead, we should expect continued moments where investor sentiment overvalues or undervalues shifting business and technological realities,” hold both Wellington Management experts.
Their position is that investing in the constantly evolving AI universe requires not only stock selection, robust analytical capabilities, and top-tier active management, but also a comprehensive understanding of its ecosystem, which can offer investors greater composure amid market volatility and ambiguity. “Simply diversifying across a basket of AI-exposed stocks is unlikely to capture the full potential of this unique and transformative technology. Conversely, active managers with strong analytical capabilities who recognize that leadership will rotate as technology evolves and market conditions change will be better positioned to generate returns and manage risk in this new AI era,” they argue.
Furthermore, for investors, it is becoming increasingly difficult to avoid tech megacaps altogether, given their weight in global equity benchmarks and their critical role in driving productivity, innovation, and economic growth. However, “investors do not need to concentrate their exposure in a handful of U.S. large-cap companies to participate in long-term digitization and artificial intelligence trends,” warns Yan Taw Boon, Head of Thematic Strategies for Asia at Neuberger.
In his view, one of the most important current developments is that the AI infrastructure boom is broadening beyond technology itself. “Capital is increasingly flowing into sectors such as energy, construction, industrial automation, and the manufacturing of specialized components required to build and operate AI data centers. This creates a broader set of opportunities for investors seeking exposure to AI-driven growth while reducing reliance on a small group of dominant tech stocks,” the Neuberger expert explains.
For Taw, diversifying exposure across geographies, sectors, and market capitalizations will be essential so that “investors can participate in the structural growth of the tech sector while mitigating concentration risk.”
