Anthropic's Opus 4.7 Forces the AI Race Into the Open

A $14.9 million prediction market now prices Anthropic as the clear leader in artificial intelligence, as the industry's human costs mount.

Anthropic released Claude Opus 4.7 on April 16, its most capable model to date and a direct challenge to OpenAI's dominance of the AI frontier. Within days, the consequences showed up where financial media has yet to look: a Polymarket contract worth $14.9 million in open interest that asks which company will have the best AI model by the end of April.

Anthropic now sits at 69.5% on that market as of April 22. OpenAI sits at 28.1%.

The shift is not abstract brand sentiment. Opus 4.7 landed with measurable gains across reasoning, code generation, and long-context tasks, the benchmarks that enterprise buyers and developers use to choose which model to build on. Amazon deepened its strategic partnership with Anthropic on April 20, expanding cloud compute and distribution access through AWS. That deal reinforces Anthropic's infrastructure position at exactly the moment its product is pulling ahead.

OpenAI retains a loyal user base and significant mindshare, but the prediction market is pricing something specific: that Anthropic's latest release is technically superior right now, during the evaluation window that matters. The $226,000 in 24-hour volume on the OpenAI side of the contract suggests active trading, not passive acceptance. Bettors are engaging with the question and still choosing Anthropic.

Then there is Meta, priced at 0.1% on the same market. On the same day that number appeared, Computerworld reported that Meta has begun tracking employee keystrokes and screen activity to train its AI agents. The company is not competing at the frontier. It is harvesting internal data from its own workforce to try to catch up. Prediction markets have rendered that verdict with brutal clarity, and the contrast between Meta's ambition and its market standing tells a story that financial coverage has largely missed.

The real-world stakes extend well beyond corporate rankings. The New York Times reported on April 21 that AI is already eliminating jobs on Wall Street, with banks accelerating automation of research, compliance, and junior analyst functions. Forbes followed with a report on April 18 documenting 8,000 tech layoffs tied directly to companies pushing AI-first strategies, with Meta's restructuring among the largest contributors.

These are not hypothetical consequences. They are happening in the same week that the model race is being decided. The companies building the most powerful AI systems are simultaneously cutting the humans those systems are designed to replace. DeepSeek, the Chinese lab that briefly disrupted the market in January, has faded to 2.1% on the April contract. The race has consolidated into a two-horse contest, and one horse is pulling away.

For investors, enterprise buyers, and workers watching from inside these companies, the signal is straightforward. The prediction market is not guessing which company has better marketing or a larger user count. It is pricing which model is best right now, with real money, and the answer is Anthropic by a wide margin. Whether that lead holds through the summer depends on OpenAI's next release cycle and whether Anthropic can convert technical superiority into the kind of enterprise adoption that generates durable revenue.

The next inflection point to watch: OpenAI's expected model refresh in May or June. If Anthropic still holds above 60% on similar markets after that release, the lead may be structural rather than cyclical. For now, the money has spoken.