work-i-love
Max Tegmark
Life 3.0 and research on intelligence, AI safety, and the limits of oversight.
Max Tegmark’s research group at MIT connects work on intelligence with AI safety. His book Life 3.0 was an important starting point for my interest in responsible AI.
Research I want to explore further includes the coauthored papers Towards Guaranteed Safe AI and A Decision-Theoretic Formalisation of Steganography With Applications to LLM Monitoring. They raise useful questions about the assumptions behind safety guarantees and the evidence needed for effective oversight.