Jamie Dimon warned this week that markets may be showing “too much exuberance,” singling out frothy valuations around artificial intelligence and the Big Tech giants building the infrastructure behind it.
Dimon made the remarks in a TV interview, saying “They could be much higher than they are today,” and that “Bond rates can go up,” adding “The notion that somehow people say they will never go up is the wrong notion.” He linked market risk to fiscal size, noting “US government debt is $30 trillion, the average rate is 3.5%,” and that “They have another $2 trillion to do this year but the thing is we don’t know when — we don’t know when the world gets too scared about that, when inflation makes it where people don’t want to own long-term duration securities.” He warned plainly that “Rates can easily go up more, and credit spreads can go up more,” and that market participants should prepare: “Companies like us prepare for higher rates, lower rates.”
The warning followed a separate note from Panmure Liberum strategist Joachim Klement, who wrote about “irrational exuberance” in the AI boom about a week before Dimon’s interview and added bluntly that “AI is in a bubble.” Panmure’s team calculated that the current AI boom is already about 60% larger than the late-1990s TMT episode when measured by the contribution of tech capex to U.S. GDP growth, and estimated that almost all of U.S. real GDP growth right now is being driven by technology investment.
Those figures give the latest comments weight: a major bank chief warning of excess, a strategist calling the rally a bubble, and a research team saying the technology investment cycle is larger than the comparable late-1990s surge.
The language recalls Alan Greenspan’s December 1996 speech at the American Enterprise Institute, when he coined the phrase “irrational exuberance” and asked, “How do we know when irrational exuberance has unduly escalated asset values, which then become subject to unexpected and prolonged contractions?” Greenspan’s remark provoked markets around the world to sell off, though U.S. stocks — especially tech names — continued to surge for several more years before the dot-com bust. Greenspan also cautioned that “we as central bankers need not be concerned if a collapsing financial asset bubble does not threaten to impair the real economy, its production, jobs, and price stability.”
The tension is immediate. Dimon’s fiscal and interest-rate warnings emphasize a path by which a valuation shock could translate into broader economic pain. Klement and Panmure frame the issue as a balance of scale and timing: “AI is in a bubble,” Klement wrote, but he allowed it could last another “one to two years.” That gap — a high, large bubble that might keep inflating for another one to two years — mirrors the Greenspan episode, when a stark warning produced an initial sell-off but did not stop a multi-year climb in stocks.
The clash highlights the policy problem Greenspan posed three decades ago: how to tell when speculative fever is simply market noise and when it has “unduly escalated asset values” enough to threaten the real economy. Dimon’s comments — and Panmure’s calculations that the current cycle is roughly 60% larger than the late-1990s TMT episode — sharpen that old question for a new, AI-driven expansion.
“How do we know when irrational exuberance has unduly escalated asset values, which then become subject to unexpected and prolonged contractions?”
