Research
CAIA members accepted to ICML 2026 workshops
CAIA members contributed to multiple papers accepted to ICML 2026 workshops, spanning pluralistic alignment, trustworthy AI, and mechanistic interpretability.
▎
We conduct research and outreach to advance the development of safe AI.
Managing risks from advanced AI is one of the most important challenges of our time.
CAIA is a community of student technical and policy researchers at Cornell working to reduce these risks and improve the trajectory of AI development.
We run an introduction to AI alignment fellowship covering topics like neural network interpretability, learning from human feedback, goal misgeneralization, eliciting latent knowledge, and evaluating dangerous capabilities in models. Interested students can learn more on the programs page.
We also run an intermediate technical reading group, support undergraduate and graduate students in original research, and host workshops and socials.
Managing risks from advanced artificial intelligence is an urgent global problem1. If you want to get involved, start by joining our mailing list.
Upcoming Events
Join talks, reading sessions, and workshops from the CAIA community.
Subscribe to get events directly on your calendar.
Subscribe to Events CalendarNews
Most recent updates from Cornell AI Alignment.
Research
EigenBench accepted to ICLR 2026 as an Oral paper!
EigenBench was accepted to ICLR 2026 and selected for an Oral presentation.
Read updateAnnouncement
RAISE Act is signed by the NY Governor
Governor Kathy Hochul signed the RAISE Act into law, establishing nation-leading AI safety requirements for frontier model developers. CAIA helped canvass support for the bill.
Read updateOur members have worked with:
