Welcome to the group of people interested in technical AI safety research. We've created this group so that everyone can access the information about how to get into the field.

Contacts:

Start here:

Diving into AI safety research:
- An in-depth look at the problem with suggestions for getting more involved
- AGI Safety Fundamentals technical alignment curriculum – 101, 201
- How to pursue a career in technical AI alignment, an outline of the various types of alignment work, and roadmaps to pursue them
- Levelling Up in AI Safety Research Engineering, a comprehensive guide to building crucial engineering skills
Key texts motivating the problem:
- Superintelligence, the definitive introduction to the area (a little dated now)
- Human Compatible, a slightly more technical look into the problem
- The Alignment Problem, crystallised from numerous interviews with AI safety researchers
- (Click here for a free copy of any of the above)
Active venues of research:
- Shallow review of live agendas in alignment & safety (Nov 2023)
- What everyone in technical alignment is doing and why (Aug 2022)
Learn more:
- AI Safety Support’s Lots of Link

Some more resources listed here (with some redundancy with the above).

AI Safety Google Group (previously 80,000 Hours AI Safety group)

Description

Language

Privacy