The Tech Capitulation Pool

This Pool was founded on the conviction that current AI valuations — embedded in the market caps of companies whose primary narrative is AI-driven revenue transformation — reflect overestimated near-term productivity gains and underestimated regulatory friction. We are not anti-AI. We hold no view on whether artificial general intelligence is achievable, or whether the long arc of AI development is positive. We hold a specific, bounded view: that the earnings revisions implied by current market pricing will not arrive on schedule. The gap between what is being priced and what will be demonstrably true in the next two to three earnings cycles is the Pool's thesis. Several data points anchor this conviction. Enterprise software adoption curves, historically, take three to five years to move from pilot to meaningful P&L impact — even for transformative technologies. Generative AI tools have been widely available to enterprises since early 2023. The productivity gains showing up in earnings are, so far, real but modest. The multiples suggest something categorically larger is already priced in. We think the adjustment happens before the consensus admits it. We also hold a view on the regulatory timeline. Frontier AI models are increasingly under legislative scrutiny in the EU, the UK, and — more slowly — the U.S. The cost of compliance, the pace of capability development constrained by safety requirements, and the liability landscape for AI-generated outputs all represent headwinds that are not adequately reflected in current market pricing of AI-exposed equities. The Pool holds positions in markets where implied probabilities of AI-driven productivity displacement appear too high, and in markets where regulatory friction appears underpriced. We are backed by the conviction that prediction markets, unlike equity markets, resolve on precise binary conditions — which is where our edge lies.