TruthfulAI

Videos

A collection of talks, interviews, and presentations about our AI safety research.

Owain Evans - Weird Generalizations and Backdoors: New Ways to Corrupt LLMs

Weird Generalizations and Backdoors: New Ways to Corrupt LLMs

The Hinton Lectures 2025 - AI Agents: Risks and Opportunities (Night 1) | Owain Evans

The Hinton Lectures 2025 - AI Agents: Risks and Opportunities (Night 1)

Owain Evans on LLM Psychology

Owain Evans on LLM Psychology

Owain Evans - Emergent Misalignment

Emergent Misalignment

Owain Evans - Deluding AIs

Deluding AIs

Owain Evans | Truthful language models and AI alignment

Truthful Language Models and AI Alignment