Artificial intelligence models are rapidly increasing, intensifying competition. With so many participants, determining which will emerge as the best—and who makes that call—is challenging. Arena, originally LM Arena, has quickly become the leading public leaderboard for cutting-edge LLMs, impacting funding, launches, and public relations. In just seven months, the startup evolved from a UC Berkeley PhD research project to a valuation of $1.7 billion.
Join Equity’s host Rebecca Bellan as she converses with Arena co-founders Anastasios Angelopoulos and Wei-Lin Chiang about the platform’s rise to prominence as the top choice for AI model leaderboards and the efforts to maintain a neutral benchmark despite support from major companies like OpenAI, Google, and Anthropic.
They explain Arena’s unique operation, highlighting its resistance to manipulation compared to traditional benchmarks, delve into the concept of “structural neutrality,” discuss Claude’s current leading position in legal and medical expert leaderboards, and outline the firm’s expansion to evaluating coding and real-world tasks through a new enterprise product.
Subscribe to Equity on YouTube, Apple Podcasts, Overcast, Spotify, and other platforms. Follow Equity on X and Threads at @EquityPod.