Former US Securities and Exchange Commission chair Gary Gensler says the AI boom is a bubble that could burst before the 2028 election and trigger a recession. His argument is that infrastructure spending is racing ahead of the revenue needed to justify it, though another market veteran quoted by CNN sees a possible correction rather than a crash.
Watch Desk analysis
What happened
In an interview published by CNN on 16 November, Gensler pointed to estimates of about $800 billion a year going into AI infrastructure. He said major technology companies are spending five times more in 2026 than they were three years ago, and warned that heavy investment, debt and complex financing could leave companies exposed if expected returns fail to arrive.
Gensler identified specialist cloud-computing providers, or neoclouds, as especially vulnerable to a downturn because some have built their businesses on substantial debt. He expects any failures or bankruptcies to land there before they threaten the biggest technology companies. These are his forecasts, not established outcomes.
CNN also interviewed Ed Yardeni, president of Yardeni Research, who argued that bubble concerns may help cool excessive investment without causing a full-blown crash. He said current chipmaker valuations do not look like the dot-com era’s peak.
Why it matters
The AI build-out is not just a debate about whether the technology works. It is also about who is financing the data centres and chips, how quickly that investment can earn money, and where losses would land if expectations fall short. Gensler’s warning puts indebted specialist cloud providers in the frame; Yardeni’s counterargument is that market scepticism may itself limit the downside.
The disagreement is useful precisely because neither side can settle the future with a confident metaphor. The scale of spending makes the economic stakes significant, but a forecast of a recession is not evidence that one is inevitable.
Our read
Treat this as a warning to examine the financing and business models behind the AI boom, not a countdown to a crash. Gensler offers a pointed case for why debt-heavy infrastructure operators could be exposed; Yardeni offers a reason not to mistake a market wobble for a bursting bubble. The interesting test is whether real demand catches up with the build-out before investors lose patience.
What to watch
- Whether AI infrastructure spending translates into durable revenue.
- How debt-heavy specialist cloud providers fare if demand or financing weakens.
- Whether company earnings and market valuations support either forecast.
Discussion spark: If AI infrastructure demand grows but the companies financing it still lose heavily, would you call the boom a success or a bubble?
Sources and evidence
- AI bubble will likely burst and cause a recession by 2028 election, Wall Street’s former top cop says – CNN (16 November 2026, 10:00 UTC)
Watch Desk is operated by WittyWires as an independent cross-cutting AI news tracker. It does not speak for the organisations or people it covers.