All Recordings

Events overview

Browse on YouTube

Thank you! Your submission has been received!

Oops! Something went wrong while submitting the form.

Dangerous Capability Evals: Basis for Frontier Safety

Mary Phuong

07/2024

Model Evaluation

Mechanistic Interpretability: A Whirlwind Tour

Neel Nanda

07/2024

Interpretability

Some Lessons from Adversarial Machine Learning

Nicholas Carlini

07/2024

Robustness

Scaling Reinforcement Learning from Human Feedback

Jan Leike

07/2024

Governance for Advanced General-Purpose AI

Helen Toner

07/2024

Governance

Stress-Testing Capability Elicitation

Dmitrii Krasheninnikov

07/2024

Resilience and Interpretability

David Bau

07/2024

Interpretability

Open Problems in Technical AI Governance

Ben Bucknall

07/2024

Governance

Measuring and Improving Human Agency in a World of AI Agents

Alex Tamkin

07/2024

Agents

Towards Reliable Alignment: Uncertainty-Aware RLHF

Aditya Gopalan

07/2024

Modeling and Mitigating Near-term Deployment Risks from LLMs

Alex Pan

07/2024

Simplex

Multiple Speakers

06/2024

Formal Methods

Formal AI-Assisted Code Specification and Synthesis

Shaowei Lin

05/2024

Formal Methods

Defending Against Adversarial Attacks in Go

Tom Tseng

05/2024

Robustness

Category Theory

Kris Brown

04/2024

Formal Methods

How Could We Design Aligned & Provably Safe Al?

Yoshua Bengio

04/2024

Formal Methods

Sleeper Agents

Ethan Perez

02/2024

Agents

Towards Quantitative Safety Guarantees and Alignment

Yoshua Bengio

12/2023

Formal Methods

Challenges for AI Safety in Adversarial Settings

Dawn Song

12/2023

Adversarial Attacks on Aligned Language Models

Zico Kolter

12/2023

Robustness

What (and Why) is a Linear Representation?

Victor Veitch

12/2023

Interpretability

Cognitive Dissonance: Why do Language Model Outputs Disagree

Stephen Casper

12/2023

Interpretability

Foundations of Cooperative AI

Vincent Conitzer

12/2023

Governance

Emergence of Complex Skills in LLMs

Sanjeev Arora

12/2023

Interpretability

System Two Safety

Shane Legg

12/2023

Oversight

Research

Our research explores a portfolio
of high-potential agendas.

Events

Our events bring together
global leaders in AI.

Programs

Our programs build the field of trustworthy and secure AI