AI Governance and Strategy

Research Sprints

View the projects submitted to our AI governance and strategy research sprints. Participants were given just 6-8 hours to select a topic and produce a presentation and research summary.
If you would like to ask any questions or be connected to any participants contact us.

March 2024 Projects

For our fourth iteration of our AI Governance research sprint we provided an updated list of research topics and questions with Nathan Barnard. The topics included: AI regulation and other standard tools, compute governance, corporate governance, international governance, misuse, evals, China, information security, strategy and forecasting, post TAI/ASI/AGI governance.

  • Winning Project: Evals as audits: lessons from the literature on auditor-firm separation

    James Lester, Theo Cochrane, Sergi Lange-Soler
    Presentation

    Summary

  • How can we ensure that evals are not gamed? An analysis drawing on the Volkswagen gas emissions test case study

    Rebecca Hawkins, Vinay Hiremath, and David Varga
    Presentation

    Summary

  • China's Potential to Establish a State-of-the-Art AI Chip Fab by 2027

    Zach Liu, Theo Kitsberg, and Gabor Fuisz
    Presentation

    Summary

  • “Digital friends” - Exploring Regulation of AI social interaction

    Naomi Colvin & Belle Yeung

    Summary

  • The Value of Corporate Governance for Improved AI Safety

    Eleos Citrini, Joe Fennell, & Omer Bilgin
    Presentation

    Summary

November 2023 Projects

In November 2023 we ran a research sprint on AI alignment and AI governance (loosely) inspired by Holden Karnofsky’s list of important, actionable questions for the most important century.

  • Winning project: Designing Incident Reporting Processes for Frontier AI

    By Matthew Wearden & Ketana Krishna
    Presentation

    Summary

  • EU AI Liability Directive

    Kate Chernyshova, Giorgio Scolozzi, Omer Bilgin, Flora Navares
    Presentation
    Further notes

    Summary

  • Proposition for a Scientific Institution for the Development of Frontier AI

    Caitlin Amy Russell & Sofia Surla
    Project submission
    Presentation

    Summary

  • Timelines & Takeoff Speeds (How have AGI timelines changed over time?)

    Rusheb Shah, Alfie Lamerton, Cian Mullarkey & Leo Falcomer-Dawson
    Presentation

    Summary

  • What are the plausible US legal outcomes surrounding AI training data by 2025? A scenario-based modelling approach

    Rong Zhou, Charlie Harrison, Maham Saleem & Ayushmaan Sharma
    Presentation

    Summary