TS Lombard scenario analysis: An AI bubble burst could trigger a 30% plunge in the S&P 500, wiping out 20 trillion dollars in market value—but don't rush to short!

According to Golden Ten Data APP, Dario Perkins from TS Lombard predicted in a scenario analysis released on Thursday that if AI capital spending collapses, roughly $20 trillion in market value would be wiped out from the US stock market, triggering $125 billion to $190 billion in credit losses and leading to a mild recession—but investors can protect themselves by diversifying global stock allocations and strategically buying bonds.

Perkins simulated the financial shock of a sudden drop in AI capital spending, forecasting that the S&P 500 index would fall 30%, tech stocks would decline by 40%, and other sectors would drop by 20%.

“Market concentration means that a slowdown in AI-related spending will deal a heavy blow to the US stock market,” Perkins wrote. “This will trigger substantial losses in the tech sector and apply credit pressure.”

The credit losses will be concentrated in the weaker links of the AI ecosystem—new cloud service providers such as CoreWeave, independent model developers like OpenAI and Anthropic, and building suppliers—rather than the ultra-large-scale cloud providers. The latter have strong balance sheets and should withstand the shock. By the end of 2027, the AI ecosystem among non-ultra-large-scale providers is expected to hold about $545 billion in outstanding bonds, $920 billion in bank loans, and $650 billion in private credit exposure.

Despite the astonishing destruction of wealth, Perkins expects only a mild recession similar to those in 1990 or 2001. The end of the AI data center boom will reduce US GDP by about 2 percentage points and cut around 800 thousand construction-related jobs. “But in terms of recession, we don't think this one will be particularly severe,” he noted, citing healthy private sector balance sheets and ample room for rate cuts by the Federal Reserve.

On Thursday, renewed concerns about the sustainability of AI, rising costs, and commercial returns prompted investors to exit stocks—especially tech stocks—and seek safety in government bonds. This followed warning signals in the earnings reports from Meta Platforms and IBM.

For investors facing such uncertainty, Perkins recommends patience and diversification, rather than aggressive shorting. “We don't know exactly when this wave of AI capital spending will end,” he admitted. “Investors who short US tech stocks (or extend duration) too early will certainly underperform.”

The recommended strategy is to maintain normal risk exposure through a diversified international stock portfolio—if the boom continues, this portfolio should perform reasonably well; if it crashes, it should outperform. “This bubble is different from the internet bubble because it is more concentrated in the US and the AI supply chain,” Perkins explained, noting that non-US markets haven't been subjected to concentrated bets on a single theme.

The report suggests that government bonds should “only be increased after signs of fragility emerge;” as the Federal Reserve responds to the demand shock triggered by AI, long-term rates are expected to decline significantly. Meanwhile, the dollar is “ultimately expected to weaken,” as the damage is focused on the US financial system.