Organisation
MIRI
One of the oldest AI-risk organisations, focused on the theory of aligning advanced AI and, more recently, on advocacy for slowing frontier development.
The Machine Intelligence Research Institute (MIRI) is among the oldest organisations working on risks from advanced AI, founded in 2000 as the Singularity Institute and renamed in 2013. Based in Berkeley and long associated with co-founder Eliezer Yudkowsky, it helped originate the field of AI alignment and the argument that a sufficiently capable AI could pose an existential threat. MIRI's early work was largely mathematical — attempts to formalise how one might build a trustworthy powerful agent — but in more recent years it has shifted much of its effort toward public advocacy for slowing or halting frontier development, a stance more pessimistic than that of most other safety groups. Its influence rests less on current research output than on having shaped how a generation of researchers frames the problem.
- Category
- Safety & alignment research
- Founded
- 2000
- HQ
- Berkeley, US
- Key people
- Eliezer Yudkowsky, Nate Soares
Appears alongside
Featured in threads
Tracks
- Ideas & essays 2
Yudkowsky and Soares publish 'If Anyone Builds It, Everyone Dies'
The authors, who had argued the case for two decades within the field, called for a global halt to large-scale AI development; reviewers split sharply on whether the argument held.
Ideas & essays
Yudkowsky publishes 'AGI Ruin: A List of Lethalities'
Yudkowsky's 43-point case that alignment is 'lethally difficult' argued no known research path could make a first critical AGI attempt survivable.
Ideas & essays