Machine Intelligence Research Institute
Ensuring smarter-than-human AI has a positive impact through foundational mathematical research on AI alignment.
Year Founded:
2000
Geography:
North America
Address:
548 Market Street PMB 46597 San Francisco, CA 94104, USA
Visit WebsiteTo ensure that the creation of smarter-than-human intelligence has a positive impact. MIRI conducts foundational mathematical research aimed at ensuring that AI systems are safe and beneficial.
AI areas they serve
Areas of Focus
AI alignment
Agent foundations
Decision theory
AI safety mathematics
Corrigibility
Value alignment
Existential risk from AI
Foundational AI safety research
AI Impact Areas
Upcoming Goals
Continue foundational research on AI alignment
Develop mathematical frameworks for safe AI
Advance understanding of alignment challenges
Contribute to the growing AI safety research community
2013: Renamed to Machine Intelligence Research Institute
Pioneered technical AI alignment research including work on corrigibility, decision theory, and agent foundations
Published foundational papers on AI alignment including Yudkowsky's "Intelligence Explosion Microeconomics”
Hosted multiple AI safety workshops
Influenced the broader AI safety research agenda
2000: Founded as the Singularity Institute for Artificial Intelligence by Eliezer Yudkowsky