Donald Trump is reportedly considering appointing an ‘AI czar’ to coordinate federal artificial intelligence policy. In ordinary times, increased government oversight might be expected to tamper irrational exuberance. But these are not ordinary times. As the recent crypto rally perhaps suggests, even perceived support from the incoming Commander in Chief is enough to drive prices through the roof. But just because Trump’s imprimatur might be sufficient to inflate a bubble, it doesn’t necessarily follow that the current AI hype cycle is a bubble. After all, AI has multiple clear use cases, and promises to be a truly transformative technology. On the other hand, the internet changed the world forever, but you wouldn’t have wanted to be on the wrong side of the dot com bubble.  

Concentration concerns 

In its Financial Stability Review published in November 2024, the European Central Bank warned about risk concentration, which could “render financial markets susceptible to sudden, sharp adjustments, notably in equity markets”. 

The ECB flags “concentration of equity market capitalisation and earnings among a handful of single names” – particularly in the United States, specifically AI firms.  

The reliance on a relatively small group of companies creates risk: 

“This concentration among a few large firms raises concerns over the possibility of an AI-related asset price bubble. Also, in a context of deeply integrated global equity markets, it points to the risk of adverse global spillovers, should earnings expectations for these firms be disappointed.” 

Notably, the risk isn’t dependent on any specific doubts about AI companies’ paths to profitability. Rather, such concentration itself poses a risk: poor results from a handful of companies could reverberate around the world.  

Mind your language model 

AI has enormous promise, but will it deliver? The real-world evidence so far is mixed. In a piece published by Harvard’s Allen Lab for Democracy Renovation, researchers raise several reasons to moderate our expectation.  

Among them, there is no obvious way to stop large-language models (LLMs) from producing errors. That’s because these models are probabilistic, rather than ‘understanding’ text in the way we traditionally think of it. 

Moreover, the authors point out, CEOs that have so far deployed AI in their companies have often seen a decrease in productivity, not an increase. 

With these issues in mind, and the reality in increasing interest rates, the researchers suggest a degree of scepticism is warranted. 

The hype is real? 

A note from Goldman Sachs is more sanguine. Unambiguously entitled ‘AI stocks aren’t in a bubble’, the piece observes that many of today’s leading AI firms were already dominant in the previous wave of tech innovation (especially cloud services and software).  

Goldman’s research suggests that, while these companies could accelerate their growth, we could see additional competition, with a new group of AI star companies emerging. (For an alternative perspective, see Goldman’s report ‘Gen AI: Too Much Spend, Too Little Benefit?’) 

And while the bank recognises the risk of market concentration (pointing out the need for investors to diversify their portfolios) they point out that concentration in this case may not be irrational. After all, the massive amounts of capital needed to compete on the AI space exclude small businesses. 

Arguably, the less headline-catching but more substantive AI technologies are only starting to be rolled out. If that’s true then, as one strategist puts it, perhaps the real AI revolution has just begun. 

For more insight into how AI could impact our future, check out The Watchlist 2025, where we unpack the trends shaping our future. 

All opinions, news, research, analysis, prices or other information is provided as general market commentary and not as investment advice and all potential results discussed are not guaranteed to be achieved. The information may have been derived from publicly available sources, company reports, personal research, or surveys. Past performance is not indicative of future performance. Trading carries risk of capital loss. Service available to professional clients only.