News

CAIA open house & mocktail mixer

We welcomed new and returning members for mocktails and conversation at our open house. Thank you to everyone who came to meet the community and learn about getting involved in AI safety at Cornell.

Guest talk with Prof. Paul Gölz

Prof. Paul Gölz joined CAIA to present Distortion of AI Alignment: Does Preference Optimization Optimize for Preferences? The talk explored how preference optimization can distort diverse human preferences and what this means for pluralistic AI alignment. Thank you to Prof. Gölz and everyone who joined the discussion.

• Talk slides (PDF) (opens in a new tab)

CAIA papers accepted to ICML 2026 workshops

CAIA members contributed to multiple papers accepted to ICML 2026 workshops, spanning pluralistic alignment, trustworthy AI, and mechanistic interpretability.

• Pluralistic Preference Alignment (opens in a new tab)
• Evaluation Awareness in Browser-Agent Safety Benchmarks (opens in a new tab)
• Side Effects of Character Training (opens in a new tab)
• Localizing Neural Causal Abstractions (opens in a new tab)

EigenBench selected for an ICLR 2026 oral presentation

EigenBench was accepted to ICLR 2026 and selected for an Oral presentation.

New York signs the RAISE Act into law

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.

CAIA ice skate social

Thank you everyone for a great semester working on AI safety. We wrapped up with a fun CAIA ice skate social.

CAIA mini-hackathon

Thank you to everyone who attended our workshop and mini-hackathon today.

• Slides (opens in a new tab)
• Starter code (opens in a new tab)
• Solution (opens in a new tab)

CAIA bowling social

CAIA members gathered for an evening of bowling at Helen Newman Hall.