The “You Can’t Play” leaderboard is funded by the companies that rank it

AI models are proliferating rapidly, and competition is intense. With so many players crowding the space, who will be the best – and who will decide? Arena, formerly known as LM Arena, has emerged as the de facto public leaderboard for frontier LLMs, influencing funding, launches, and PR cycles. In just seven months, the startup has gone from a PhD research project at UC Berkeley to being valued at $1.7 billion.

Watch as Equity host Rebecca Bellan connects with Arena founders Anastasios Angelopoulos and Wei-Lin Chiang about how their platform has become the leaderboard of choice for frontier AI models, and how they’re trying to build a neutral standard even as companies like OpenAI, Google, and Anthropic support the project.

It explains how Arena works and why it’s harder to play than static benchmarks, what “structural neutrality” actually means, why Claude currently tops expert leaderboards in legal and medical use cases, and how the company is expanding beyond chat to real-world benchmarking, programming, and tasking agents with a new enterprise product.

Subscribe to Equity on YouTube, Apple Podcasts, Overcast, Spotify, and all the cast. You can also follow stocks on X and topics, in EquityPod.