Paul Riechers
Simplex AI safety

Paul Riechers is a theoretical physicist with expertise in the physics of information, and the ultimate limits of learning and inference. He co-founded and leads AI interpretability research at Simplex, investigating internal representations and emergent behaviors of AI. He is concurrently a Senior Fellow at UCLA’s Mathematics of Intelligences program at the Institute for Pure and Applied Math (IPAM). As co-founder and Senior Scientist at the Beyond Institute for Theoretical Science (BITS), he seeks to understand the trajectory and limits of intelligence in our universe.

Society is on track to build superhuman general intelligence in the near future, whether or not we understand it. By default, powerful AI systems will pursue intentions formulated with respect to an alien worldview, sometimes conflicting with basic human priorities. To enable human flourishing, significant resources must be invested now to determine and steer the interrelation of concepts that AI systems internally represent and utilize.