The Warring States Period: Frontier Labs Edition

LessWrong is reporting a major shift in frontier AI: the race has gone from what looked like a two-company battle between Anthropic and OpenAI to a broader oligopoly, with SpaceX's Grok four point five, Meta's Muse Spark one point one, and Alphabet's Gemini models all back in serious contention. Anthropic still leads by revenue—estimated at over sixty billion dollars annualized—but the analysis flags growing trust risks around data handling and customer access. OpenAI, conversely, is being underestimated on its consumer distribution and advertising potential. The key strategic question: without one lab achieving a decisive breakthrough in self-improving AI systems, model leadership will stay contested. That means the real battle will shift from pure capability to conventional moats—distribution networks, cost, and customer lock-in. As a wild card, SpaceX has found a near-term revenue stream renting surplus compute to Anthropic and Alphabet, generating rates around fifty to seventy-four billion dollars per gigawatt annually—unsustainable, perhaps, but indicative of how capital-intensive and fluid this competition has become.

Listen to this story

Hear this and more stories in a personalized audio briefing.

Open The Chonkerton