June 2026

The meteoric rise of generative artificial intelligence has undoubtedly been the defining market narrative of the 2020s. However, as capital continues to flood into AI infrastructure and tech megacaps achieve unprecedented valuations, a familiar and uneasy whisper is growing louder across trading floors and boardrooms: Are we in a bubble?

Recent market commentary has increasingly drawn parallels between the current AI frenzy and the Dot-Com Bubble of the late 1990s. The similarities are hard to ignore. Just as the commercialisation of the internet triggered a speculative gold rush in 1999 – where any company adding “.com” to its name saw its stock soar – today’s market exhibits a similar fervor for anything touched by machine learning. Critics warn that we are once again pricing in decades of future growth into today’s multiples, heavily front-loading capital expenditures before clear monetisation models have fully matured.

History reminds us that this script has been written before, and not just in the 1990s. Consider the Railway Mania of 1840s Britain, where thousands of miles of tracks were funded by speculative investors convinced that the locomotive would reshape commerce. It did – but not before a brutal market crash wiped out a generation of over-leveraged backers. Similarly, the Telecom Bubble of the early 2000s saw over-investment in fiber-optic cables that lay dark for years before demand finally caught up with capacity.

The core lesson from these historical cycles isn’t that the underlying technology is a fraud. The internet did change the world. Railroads did revolutionise global trade. The issue is timing and capital discipline. Financial history shows that pioneering infrastructure is almost always overbuilt during the initial hype cycle, leading to a painful correction where weak players are purged, clearing the path for sustainable, long-term deployment.

As financial services professionals, our mandate is to separate transformational utility from speculative noise. AI will fundamentally reshape productivity, risk management, and quantitative analysis. However, managing risk means questioning whether current cash flows justify present valuations, or if we are merely underwriting the hype.

Whether this turns out to be a minor market digestion or a structural correction, maintaining rigorous valuation discipline remains our best defense. What’s your perspective? Is your firm leaning into the AI momentum, or are you actively hedging against a potential correction?