Just came across an interesting take on why people worry about AI bubbles. Bloomberg made a solid point about something most people miss when comparing AI to the dot-com bubble situation.

The thing is, back during the dot-com era, regular retail investors were the ones taking the hits when everything crashed. They were putting their own money into these startups with minimal due diligence. But look at what's happening with AI funding now - it's a completely different animal. Institutional money is driving the show. We're talking about massive funds, corporations, and institutional players betting on AI infrastructure and companies.

That distinction actually matters way more than people realize. Sure, both periods have that speculative energy, that sense of hype and potential. You can feel it in the market. But the financial consequences play out differently when institutions are the primary risk-takers versus retail crowds. The scale is different, the leverage profiles are different, even how contagion spreads during corrections looks different.

So while the dot-com bubble comparison gets thrown around a lot - and there are legitimate similarities worth discussing - the investor profile shift between then and now could actually dampen some of the worst-case scenarios. Doesn't mean AI valuations can't correct or that there won't be casualties in the sector. Just means the mechanics of any potential crash might look different from what happened in 2000.

Worth keeping an eye on as this AI cycle matures.