Artificial intelligence was supposed to be the most certain investment theme this year. However, it has transformed into a threat, not to the tech giants building AI, but to asset-light companies that could be replaced by AI. This week, the S&P 500 index was on track for its worst performance since November, only rebounding after mild inflation data on Friday. The panic over AI disruption is spreading across various market segments. White-collar industries such as software firms, wealth management institutions, brokers, and tax advisors, which had seen years of margin expansion, are being repriced within weeks. The shockwaves have even reached the private credit market that lends to these companies.
Wall Street's highly confident bets have failed comprehensively over the past six weeks. Fund managers, who started the year with record-low cash allocations and minimal hedging, are now witnessing the collapse of consensus trades. The most favored assets are losing to the most unpopular ones. Energy, consumer staples, and U.S. Treasuries are leading the market in 2026, while the consensus AI bets from the start of the year are faltering. The iShares 20+ Year Treasury Bond ETF (TLT) recorded its largest weekly gain since April, while the SPDR S&P 500 ETF Trust (SPY) has lagged behind TLT by 2 percentage points since December, marking its worst start to a year in a decade.
The AI theme, initially seen as a surefire opportunity, is now becoming the market's greatest source of uncertainty. Investors are beginning to question the timeline for returns on the massive capital expenditures by tech giants and whether remaining cash can continue to support stock buybacks. Adam Crisafulli, founder of Vital Knowledge, stated that over the past few months, AI has hurt more stocks than it has helped. Jim Caron, Chief Investment Officer at Morgan Stanley, noted that the market is experiencing a repricing in a specific sector, namely software. There is concern this could trigger a contagion event spreading to other areas. He is focused on two questions: whether AI-induced losses will cause contagion, and how to hedge this risk through diversification.
Two forces are amplifying volatility in the U.S. stock market. The first is positioning. A January investor survey from Bank of America showed cash allocations fell to a record low of 3.2%, with nearly half of fund managers holding no downside protection, the lowest level since 2018. The second is the leverage network connecting seemingly unrelated portfolios, where liquidations in one corner trigger selling in another. James Athey, Portfolio Manager at Marlborough Asset Management, indicated the biggest risk is an additional volatility shock event. Everything appears highly correlated, so selling in one asset could force selling in others.
A broad decline in U.S. stocks on Thursday triggered algorithmic selling in metals, forcing some investors to exit commodity positions, including metals, for liquidity. Gold fell over 3%, dropping below $5000, while silver plunged 11%. A model designed by Jordi Visser of 22V Research shows that market linkage is soaring, even with the VIX remaining low and the S&P 500 holding above its 50-day moving average. This combination is interpreted as stress hidden beneath a calm surface. Such stress signals occurred about once a month over the past two years. This year, there have been over a dozen instances in less than two months.
The VIX briefly broke above the closely watched 20 level this week. Although the reading does not indicate panic, the skew for put options remains at historically high levels, suggesting the market is systematically paying for downside protection. An ETF tracking investment-grade bonds (LQD) recorded its best weekly performance relative to a high-yield bond ETF (HYG) since October, widening its lead for the year. The yield on the 10-year U.S. Treasury note ended the week at a two-month low.
For now, sharp volatility has not turned into a sustained market crash. The S&P 500 remains near all-time highs, and credit spreads are still around decade lows. However, hedging activity is increasing, as seen in the trading volume of individual stock put and call options. The CBOE put-call ratio has surged since January, rebounding from a near four-year low. An ETF tracking companies with high shareholder returns attracted $3.6 billion in new money this month, the largest among so-called smart beta funds tracked by Bloomberg.
Analysis suggests that if negative news about AI disruption pauses and volatility decreases, U.S. stocks could find support and move higher as dealer hedging flows turn more supportive. However, as Chris Hussey of Goldman Sachs noted, there is a conflict between the market consensus that AI will disrupt a wide range of economic sectors and macro data and corporate performance that show no abnormalities. Whether the group consensus will prevail, or the post-pandemic theme of sustained economic resilience will continue, with U.S. growth and corporate profits remaining robust, is a question that may take considerable time to resolve.