Co-founder & CEO, Anthropic Former Policy Director, OpenAI Co-author, Import AI newsletter
Jack Clark is a leading voice in AI policy, governance, and safety. As co-founder of Anthropic — a company focused on building safer large language models — Clark plays a central role in aligning advanced AI systems with human values and regulatory expectations. Prior to Anthropic, he served as Policy Director at OpenAI, where he helped pioneer frameworks for responsible disclosure, alignment, and collaboration with government bodies. Through his widely read Import AI newsletter, Clark has chronicled the global development of AI capabilities, risks, and oversight mechanisms. His work combines technical understanding with policy foresight, making him a key figure at the intersection of AI development and governance.
“The most powerful AI systems will not fail in obvious ways — they’ll fail in ways that seem correct until they’re not.”