AI Safety Studio

The AI Safety Studio is Mila’s hub for technical and socio‑technical AI safety research. It convenes researchers, clinicians, and policy experts to study harms from large language models and design independent safety metrics, guardrails, and interventions—such as improved safeguards for conversational agents used by youth—to reduce risks like suicide and "AI psychosis" linked to chatbot use.

Endorsements support Mila.

No endorsements yet.

Discussion

Start Date

Oct 10, 2025

End Date

-

Expected Duration

-

Funding Raised to Date

-