MIRI

The Machine Intelligence Research Institute (MIRI) is a nonprofit research organization founded in 2000 (originally as the Singularity Institute for Artificial Intelligence) dedicated to ensuring that the development of artificial general intelligence benefits humanity. The organization conducts technical research on AI alignment, decision theory, and related mathematical foundations to address existential risks from advanced AI systems.

MIRI's research agenda focuses on theoretical problems in AI safety, including agent foundations, logical uncertainty, and value alignment. The institute has published numerous papers and maintains a research community of mathematicians, computer scientists, and philosophers working on long-term AI safety challenges.

Based in Berkeley, California, MIRI has become a central institution in the AI safety field and has influenced academic and policy discussions around existential risk from artificial intelligence. The organization is supported by donations from individuals and foundations concerned with AI safety.