William L. Anderson

Research


My work focuses on multi-agent safety and technical governance. I'm particularly interested in thinking about how the safety stack will need to change as we move towards a massively multi-agent world.


Publications

  1. Lessons from Multi-Agent Safety Incidents William L. Anderson Cooperative AI Foundation blog, 2026.
  2. ORBIT: A Framework for Multi-Agent Security Evaluations Ben Hagag*, William L. Anderson*, Srija Chakraborty, Christian Schroeder de Witt NeurIPS 2026 (Under Review). * denotes equal contributions.
  3. Think Fast: Estimating No-CoT Task-Completion Time Horizons of Frontier AI Models Dewi Sid William Gould, Francis Rhys Ward, [...], William L. Anderson, [...], Ryan Greenblatt NeurIPS 2026 (Under Review). Minor contributor.
  4. Open Challenges in Multi-Agent Security: Towards Secure Systems of Interacting AI Agents Christian Schroeder de Witt, Klaudia Krawiecka*, Igor Krawczuk*, Ben Hagag*, William L. Anderson*, et al. arXiv preprint. * denotes major contributions.
  5. Architecture Matters for Multi-Agent Security Ben Hagag, William L. Anderson, Christian Schroeder de Witt, Sarah Scheffler ICML 2026. Major contributor.
  6. What Should Frontier AI Developers Disclose About Internal Deployments? Jacob Charnock*, Raja Mehta Moreno, Justin Miller, William L. Anderson ICML TAIGR 2026. Major contributor.

Talks


Awards & Fellowships