Steve Eisman Cuts AI Risk but Won’t Go Net Short — SkimNews
Get the Finance newsletter
Daily finance — markets, central banks, M&A, the prints that move money. Free.
- Steve Eisman said he has become "much more conservative" in his portfolio, shorting against many positions to reduce risk substantially while remaining not net short on AI.
- Eisman believes frontier algorithm models are exploiting fear around the technology to build a competitive regulatory moat against cost-efficient products and obscure faltering fundamentals.
- Anthropic left its latest revenue figures blank in its initial IPO filing and included few metrics for the upcoming quarter; Eisman called the disclosure "extremely negative" and believes the revenue data will show significant deceleration from the prior quarter.
- Anthropic figures put the revenue-to-equipment-expense ratio at 2.08 in the second quarter and 1.7 in the third, a change Eisman, citing an analysis he was given, described as pointing to a capital-efficiency issue.
- Eisman said he was scared to make a definitive bet against AI because technology cannot be siloed from the broader equity market and economy, even as he believes the trade is nearing a "tipping point."
- Google said strong growth in AI Mode queries brought lower monetization; after Wednesday’s earnings, Alphabet shares "tanked" following a core-advertising pullback as AI queries supplanted search, and Google said ad experiments were promising.
Why it matters: Eisman’s split between bearish analysis and portfolio positioning makes the AI trade’s market entanglement concrete: he has shorted against multiple positions and sharply cut risk, yet stays not net short because technology cannot be separated from the broader equity market. Anthropic’s ratio drop and Google’s ad pullback underpin his caution.
Ask SkimNews
