Laura Weidinger

Responsible AI Researcher & Speaker

I'm an expert on AI safety evaluation, taking a sociotechnical approach that focuses on risks to people. My goal is to develop and establish methods for early risk assessment to help make AI systems safer and create a more accountable AI innovation ecosystem.

I'm always interested in research on sociotechnical AI evaluation and collaboration opportunities. I'm also a frequent public speaker and excited to help build this field. If you'd like to get in touch, email me at lauraweidinger[at]pm[dot]me.

Keynotes and Invited Talks

I speak regularly at academic and public policy institutions. See some highlights below/ Contact me about a speaking opportunity.

Academic Talks

Keynote at ICML workshop on Humans, Algorithmic Decision-Making and Society (2024)
Keynote at ACM LREC/COLING Workshop on Safety for Conversational AI (2024)
Keynote at Open Data Science Developer Conference (2024)
ACL Big Science workshop panel with Meg Mitchell, Emily Bender and Zeerak Talat (2023)
Talks at FAccT conference, Oxford University, Queen Mary University London. CARE-AI at Guelph University, Canada, AI Safety Hub of Edinburgh University
Talks and lectures at the Alan Turing Institute, Allen Institute for AI, Max Planck Institute, Ada Lovelace Institute, Ada National College of Digital skills.
I've also spoken at industry and third sector groups such as ML Commons, the Nokia Bell Lab and the BBC.

Public Policy Talks

US White House visit on Sociotechnical AI Safety (2024)
US AI Safety Institute (2024)
UK AI Safety Institute (2024)
NIST Innovation Lab with Cathy Nelson, Doug Reynolds, Brian Anderson (2024)
NIST AI Metrology Colloquia (2024)
National Academy of the Sciences with Hanna Wallach, Rishi Bommasani, Miranda Bogen and William Isaac (2024)
Keynote at European Commission TechEthos on AI Ethics (2023)
UK AI Fringe with Deb Raji, Toby Shevlane and Rumman Chowdhury (2023)

Selected Research

My academic work is focused on Evaluating AI systems for Ethics & Safety. With co-authors, I've developed a widely used taxonomy of ethical and social harms from generative AI. I'm now leading work on red teaming and other methods for safety evaluation. In addition to some highlights below, find here a full list of my publications. In addition to leading my own research, I've also contributed to DeepMind flagship work on Gopher, Sparrow, YouTutor and Gemini.

Ethical and Social Harm from AI

Ethical and social risks of harm from Language Models (pre-print, 2021).
- Peer-reviewed, abbreviated version: Weidinger, L., Uesato, J., Rauh, M., Griffin, C., Huang, P.S., Mellor, J., Glaese, A., Cheng, M., Balle, B., Kasirzadeh, A., Biles, C., Sasha Brown, Zac Kenton, Will Hawkins, Tom Stepleton, Abeba Birhane, Lisa Anne Hendricks, Laura Rimell, William Isaac, Julia Haas, Sean Legassick, Geoffrey Irving, Iason Gabriel. 2022, June. Taxonomy of Risks posed by Language Models (FAccT 2022).

Evaluation

Sociotechnical Safety Evaluation of Generative AI Systems (pre-print, 2023)
- Peer-reviewed, abbreviated version: Rauh, M., Marchal, N., Manzini, A., Hendricks, L.A., Mateos-Garcia, J., Bergman, S., Kay, J., Griffin, C., Bariach, B. Gabriel, I., and Weidinger, L. Gaps in the Safety Evaluation of Generative AI (AIES 2024).
Holistic Safety and Responsibility Evaluations of Advanced AI Models (pre-print, 2024)
- An overview of what a comprehensive approach to AI Safety Evaluation may look like.

Red Teaming

STAR: SocioTechnical Approach to Red Teaming Language Models (pre-print, 2024)
- Peer-reviewed, abbreviated version: STAR: SocioTechnical Approach to Red Teaming Language Models (EMNLP 2024)

Alignment

Using the Veil of Ignorance to align AI systems with principles of justice. Weidinger, L., McKee, K.R., Everett, R., Huang, S., Zhu, T.O., Chadwick, M.J., Summerfield, C. and Gabriel, I. Proceedings of the National Academy of Sciences, 120(18), p.e2213709120.

Evaluation Framework

Evaluation Landscape

In the News & Media

Podcasts

BBC: Speaking of Intelligence
Deutsche Welle: Science Unscripted
The Gradient: Laura Weidinger

Documentary

The Thinking Game (2024)

Laura Weidinger

Responsible AI Researcher & Speaker

Keynotes and Invited Talks

Selected Research

In the News & Media

Get in touch at lauraweidinger[at]pm[dot]me