There isn't just one AI bubble, strategist says — there's a 'rolling sequence of bubbles' instead | Fortune
Strategist Amit Joshi, formerly chief strategist for Counterpoint at London‑based BCA Research, has reshaped the conversation about whether artificial‑intelligence investments are in a bubble.
Rolling AI Bubbles, Not One Giant Burst
Strategist Amit Joshi, formerly chief strategist for Counterpoint at London‑based BCA Research, has reshaped the conversation about whether artificial‑intelligence investments are in a bubble. In a LinkedIn post he argued that the market is not inflating a single, monolithic bubble that will eventually pop. Instead, investors are witnessing a rapid succession of bubbles that rise and fall in a rolling pattern as they repeatedly misjudge which companies or sectors will capture AI’s upside.
Sector‑by‑sector Swings
Joshi illustrated his point with a chart that shows software stocks soaring on the promise that AI would boost productivity, then tumbling when investors realized AI agents could undermine the SaaS subscription model itself. He summed up the episode with the line, “So, the software boom turned to bust.” A similar boom‑and‑bust cycle unfolded in silver, where the metal’s price surged on the belief that it was the best electrical conductor for power‑hungry data centers. Joshi warned that “on reassessment however, this could not justify a near trebling of the silver price when there are other good conductors.”
Semiconductor shares also rode a wave of optimism, driven by expectations of limitless pricing power for chipmakers. Joshi countered that investors are now seeing that chipmakers lack “moats” around their profits and warned, “Astronomical margins will crash back to earth when demand and supply equilibrate, as they ultimately must.” He predicts the semiconductor boom is unwinding, though more decline may follow.
Profit‑margin Bubble, Not Earnings Bubble
In a Fortune interview Joshi said he disagrees with his former BCA colleague Peter Berezin, who described the market as being in an earnings bubble. Joshi prefers the term “profit margin bubble,” noting that earnings are not unjustified by price‑to‑earnings ratios, but the market is asking, “How is the E high?” because margins are unusually high and may be unsustainable.
Extreme Price Discovery or Mania?
Critics might label the swings as normal price discovery—testing a thesis, finding it wrong, and correcting. Joshi argues the magnitude of the moves is far beyond ordinary adjustment. He quoted himself, “If you can make a fortune in a matter of weeks or months, and, crucially, then lose it all just as quickly or even quicker,” then added, “then that constitutes a ‘bubble.’” He believes the rapid inflations and deflations reflect a narrative contagion that briefly takes hold before moving on, affecting not only equities but also commodities like silver.
Broader Market Signals
Joshi’s view is echoed by several high‑profile figures. JPMorgan CEO Jamie Dimon has repeatedly warned of elevated valuations, and Bank of America Research’s Global Fund Manager survey listed “AI equity bubble” as the top tail risk. OpenAI CEO Sam Altman, Goldman Sachs CEO David Solomon and Amazon founder Jeff Bezos have all acknowledged that something “bubbly” is happening. Yet the bubble that many expected to pop by 2025 has persisted.
The latest earnings season added a new layer to the debate. Google posted a free‑cash‑flow negative result for the first time in its history, and Reuters calculated in late July that Microsoft, Alphabet, Amazon, Meta and Oracle were on track for capital expenditures to outpace free cash flow by 2027. The discussion now centers on whether this overspending is rational.
Mixed Signals from BCA
BCA Research sent mixed messages this year, upgrading equities in May on the premise that AI‑driven capital spending is propelling markets forward. Yet BCA strategist Juan Correa cautioned, “We suspect that we could be in the early innings of a violent blow‑off rally in AI‑related stocks.”
Looking Ahead
Joshi disaggregates the AI asset class into a sequence of rolling bubbles, explaining why no single AI‑linked selloff has triggered a market crash. When asked about the peak of AI capital expenditure, he projected it would most likely occur in late 2026 or the first half of 2027. He added that the outsized earnings gains are built on “stratospheric and unsustainable profit margins,” but he would revise his view if those margins normalized without eroding profits. The rolling nature of the bubbles, he says, has so far prevented a correlated market sell‑off, but investors must stay alert to which sector will inflate next.
Source: fortune.com · 2026-08-16