Jack Clark

Jack Clark is a co-founder and the head of policy at Anthropic, the AI safety and research company behind the Claude family of models, and one of the most influential voices shaping how governments and industry think about the risks and governance of large-scale AI systems. Before co-founding Anthropic, Clark worked as a technology journalist, including at Bloomberg, where he covered the artificial intelligence industry closely, and later served as policy director at OpenAI, giving him an unusually direct vantage point on the data and compute build-out that powered the first generation of large language models.

In January 2021, Clark left OpenAI alongside Dario and Daniela Amodei and several other colleagues to found Anthropic, built around the thesis that frontier AI systems needed to be developed with safety and interpretability as first-order engineering priorities rather than afterthoughts. As head of policy, Clark has represented the company before legislators, regulators and international bodies, arguing for evidence-based AI governance grounded in empirical measurement of model capabilities rather than speculation.

Clark is also the author of Import AI, a widely read weekly newsletter tracking developments across AI research, policy and industry that has become required reading for a large swathe of the field, including many of the researchers and policymakers who set data and AI governance agendas. The newsletter, combined with his public writing and testimony, has made him one of the clearest and most consistently cited translators of technical AI progress into policy-relevant terms.

In 2026, Clark took on leadership of the newly formed Anthropic Institute, a think tank studying the societal and economic effects of advanced AI, extending his policy work beyond Anthropic's own products into broader research on how AI and the data it is trained on and generates will reshape labour markets, institutions and governance. He has been outspoken about the risks of an AI 'intelligence explosion' proceeding without adequate oversight.

Clark's influence rests on his position at the intersection of frontier AI development and the policy apparatus now being built to govern the data, compute and models driving the AI economy. As one of the seven co-founders of a company valued in the hundreds of billions of dollars and trusted with some of the most consequential decisions about AI safety, his voice carries substantial weight in debates over how the world's AI and data infrastructure should be regulated.