Vinay Rao - Safety in Foundational Models
- Alexy Khrabrov
- 5 days ago
- 1 min read

Vinay Rao is an Advisor at Anthropic. Vinay leads Engineering, Product Management, Policy Design, and Data Science teams to ensure safe deployment of Anthropic’s AI models. Prior to joining Anthropic Vinay has worked in Trust, Safety, and Risk at YouTube, Stripe, Airbnb, and Google. Vinay holds a PhD in Engineering from Stanford.
Safety in Foundational Models
Vinay Rao's talk covers the unexpected challenges that emerge when deploying increasingly capable AI systems across modalities like language, vision, and code. Vinay will examine where things can go wrong, how teams identify and mitigate risks, and what safety actually looks like in practice—from automated evaluations and red-teaming to deployment safeguards and policy integration.
Comentários