Technical Fundamentals

An eight-week reading group on technical AI safety — reward specification, generalization, interpretability, unlearning, and career explorations — based on the BlueDot Alignment course.

Commitment:
2 hr/week sessions + ≤1 hr prep, 8 weeks
For:
Anyone curious about technical AI safety; ML experience encouraged but not required

Policy Fundamentals

An eight-week reading group on AI governance — AI harms, economic and social impacts, responsible scaling, open vs. closed source — based on the BlueDot Governance course.

Commitment:
2 hr/week sessions + ≤1 hr prep, 8 weeks
For:
Anyone interested in AI policy; a public-policy background is encouraged but not required

Technical Upskilling

Build a strong foundation in machine learning and a working overview of the major topics in AI safety, with a certificate on completion.

Commitment:
Self-paced with weekly support
For:
Members preparing to participate effectively in technical Safety Scholars sessions

Safety Scholars

Our core member group: weekly technical or policy discussions over a provided meal, with office hours, skill-building, and priority access to opportunities and projects.

Commitment:
Weekly meetings, semester-long
For:
Standout intro-program graduates and students with strong AI safety background

Details

Graduate students: email aisafetyuw@gmail.com to shadow a Safety Scholars session — invitations are extended to good fits.

Undergraduates: Safety Scholars applications open at the end of each semester. Priority goes to standout intro-program participants and students with a strong AI safety background; strong applicants are encouraged to reach out directly.

New to WAISI? Start with the WAISI at-a-glance handbook.

Upcoming Events

Nothing outside of our usual programming happening soon — check back later!