Jack Clark, Co-Founder and Head of Policy at Anthropic, has a rare ability to see around corners in the artificial intelligence world, and he’s convinced we’re underestimating what’s coming. Clark believes that within just a few years, AI systems will rival the intellectual capacity of Nobel laureates and take autonomous action in the physical world, operating everything from drones to robots. Known for helping steer the development of Claude, Anthropic’s powerful frontier model, Clark is not only shaping AI’s technological direction but also its ethical and geopolitical footprint. At Anthropic, he is focused on frontier lab security, export controls, and the delicate balance of innovation and regulation.
Originally from the UK, Clark holds a bachelor’s degree in English literature with creative writing from the University of East Anglia. He began his career as a journalist covering technology before joining OpenAI as policy director, where he helped define the conversation around responsible AI development. At Anthropic, he leads policy strategy while engaging with governments and research institutions to ensure frontier AI systems are built and deployed securely. His day-to-day involves aligning national security concerns with AI capabilities, pushing for energy and semiconductor investments, and protecting sensitive AI research from hostile state actors.
What continues to fascinate Clark is the sheer pace at which AI evolves. “AI systems can now recover from errors without you telling them that they’ve made a mistake,” he observed, a development he hadn’t expected so soon. This level of self-correction marks a fundamental shift in machine reasoning. For Clark, even simple ideas like scaling compute to improve AI performance carry explosive power, triggering fierce competition among global labs. His voice remains a critical one in shaping not just what AI can do, but what it should do in the years to come.