The leaderboard “you can’t game,” funded by the companies it ranks

Arena (formerly LM Arena), a UC Berkeley PhD research spinout, has become the dominant public leaderboard for frontier LLMs, reaching a $1.7 billion valuation in just seven months. Co-founders Anastasios Angelopoulos and Wei-Lin Chiang discuss how the platform works, why it's harder to game than static benchmarks, and what 'structural neutrality' means when the ranked companies (OpenAI, Google, Anthropic) are also backers. They also cover why Claude currently leads expert leaderboards in legal and medical domains, and Arena's expansion beyond chat to benchmark AI agents, coding, and real-world tasks via a new enterprise product.