This Billionaire Hedge Fund Manager Says These Are 'Crazy Times'—But He's Buying More AI Stocks

Paul Tudor Jones said he bought more AI stocks, describing current market conditions as 'crazy times' and warning the AI trade likely has another year or two left before a 'breathtaking correction.' He compared the AI surge to the dot-com era and to early productivity cycles, but his stance remains tactical rather than outright bullish. The article highlights continued investor enthusiasm for AI-related equities alongside growing bubble-risk concerns.

Analysis

The important signal is not that an established macro trader is bullish on AI; it’s that he’s explicitly framing the trade as a late-cycle momentum regime rather than a clean fundamentals story. That usually means near-term price discovery stays dominated by capex spend, index flows, and narrative compression, while stock selection matters less than factor exposure. In that setup, the strongest relative winners tend to be the infrastructure layer with the most visible order book and the most elastic earnings revisions, while the laggards are any “AI-adjacent” names that have rerated without a direct monetization path.

For MSFT, the setup is asymmetric: it can keep compounding because it monetizes both software and cloud AI, but the market is also already paying for a multi-year AI option. That makes upside more dependent on accelerating Azure and seat-based monetization than on generic AI enthusiasm. IBM is a different animal: it benefits from the “enterprise AI” theme but needs proof that consulting and software attach are translating into durable re-acceleration, otherwise it risks being used as a low-beta proxy and then fading when investors rotate back to the highest-convexity semis and cloud platforms.

The contrarian miss is timing. Consensus often treats AI as a secular straight line, but the more likely path is a series of 10-20% air pockets driven by supply-chain bottlenecks, capex fatigue, or a single earnings season where orders fail to exceed already-lofty expectations. The likely catalyst to break the trade is not a macro recession but an earnings miss or guidance reset from one of the hyperscalers, because that would hit the whole spend chain at once and expose how much of current demand is pre-committed rather than incremental.

Best risk/reward over the next 1-3 months is still to own the leaders but express it with defined downside. If the AI basket keeps working for another 6-12 months, the incremental gain is likely concentrated in semis and cloud infrastructure; if it stalls, the unwind will be sharp and correlation-driven. That argues for staying long exposure, but only with protection or relative-value hedges rather than outright momentum chasing.

Key Decisions for Investors

• Maintain a core long MSFT vs. short IBM pair for 1-2 quarters: MSFT has higher-quality AI monetization and better earnings durability, while IBM is more vulnerable to multiple compression if enterprise AI spend broadens but does not accelerate.
• Buy a diversified AI infrastructure basket on pullbacks over the next 2-4 weeks, but fund it with shorts in second-tier AI-adjacent laggards; the trade is to own visible capex beneficiaries while fading names trading on theme alone.
• For event risk, use MSFT call spreads 3-6 months out instead of outright stock: captures another leg of AI enthusiasm while limiting damage if the market reprices the trade after a hyperscaler guide-down.
• Set up a hedged semis expression: long AI hardware / equipment beneficiaries, short a broad tech ETF, for a 1-3 month window; if the AI cycle continues, idiosyncratic capex names outperform, and if it breaks, the short cushions factor drawdown.
• Reduce gross exposure into any post-earnings spike in the mega-cap AI leaders; the risk/reward worsens materially once positioning becomes crowded and the next catalyst is a disappointment rather than a re-rating.