Anthropic Likely Holds the AI Crown Through Summer

Anthropic's Claude has become, in the judgment of the people with real money on the line, the dominant large language model in the world — and that position appears almost certain to hold through August. The question the AI industry should be asking is what happens next.

The near-term signal is about as decisive as these markets get. Anthropic's probability of holding the top spot at the end of August sits at 96%, with OpenAI as a distant challenger at 3% and Google barely registering. The deeper Kalshi market, which handled roughly five times the trading volume of Polymarket over the past day, prices Anthropic's lead at the lower end of that range — a spread worth noting, though the direction of both venues is identical. Whatever small disagreement exists between platforms, it is a disagreement about magnitude, not direction. The money is not divided on who leads.

What makes the cluster's signal genuinely interesting is the tension it contains across time. The August outlook is near-certain; the December picture is a coin flip at 70% — meaningful but not settled. That gap is the story. It suggests the people staking money here believe Anthropic's current technical edge is real but not durable, that the frontier moves fast enough to make a summer champion look vulnerable by winter. The informed crowd — likely a mix of AI researchers, product builders, and quantitatively sophisticated retail — appears to be pricing in a world where Claude leads today but the race compresses before the year is out.

Google's position as second-best lab by August has slipped just below even odds, a subtle but telling detail. If the consensus believed Anthropic's lead was merely a temporary plateau before a broader reshuffling, you might expect Google's contender status to be firming, not softening. Instead the money appears to be saying that the field behind Anthropic is itself unsettled — OpenAI and Google trading positions in the shadow of a dominant leader, rather than one clear successor consolidating.

The 92% probability assigned to Alibaba holding the top spot among Chinese AI labs reads as a parallel story: a bifurcated global AI landscape where Chinese and Western frontier models are effectively judged in separate categories, with Alibaba's domestic dominance as secure near-term as Anthropic's global one. The two near-certainties coexist without contradiction.

What would break the market's read? A surprise model release from OpenAI or Google before August that benchmarks unambiguously above Claude would be the clearest falsifier — and the odds suggest the crowd considers that unlikely but not impossible. The year-end number tells you the crowd thinks something like that, or a rapid capability jump from an unexpected direction, is plausible enough to price in meaningful uncertainty. The pre-launch perpetual market on CASHCAT on Hyperliquid, trading at a hair above its oracle reference price with near-neutral funding, adds only speculative color here — a signal that some capital is positioning in adjacent AI-adjacent assets, though formal listing has not yet occurred and these prices carry meaningful reversal risk before any spot market opens.

For anyone building products, raising capital, or making hiring decisions around frontier AI capability, the signal is directionally clear for the next few months and genuinely uncertain beyond that. Anthropic holds the crown. Whether it keeps it past summer is, by the money's own honest accounting, still to be decided.