Select research, fellowships, conferences, and field-building initiatives I’ve undertaken in the last several months of 2026.
Research & Publications
- (Aug) Behavioural Indicators of Fault in Large Language Models
- (Sep) Evidence and Access in the OpenAI-Hugging Face Incident with Joseph O’Neill and Sofiia Lobanova - this paper formed part of the required reading for the 2026 SPAR project, Verification Evidence for Compute Governance: What Inspection Regimes Actually Detect as well as for the Compute Governance Career Accelerator Fellowship.
- (Sep) Submission to the Australian Parliament Joint Select Committee on AI
Field-building & Founding
- (Jul) What Lawyers Can Do For AI Safety - Less Wrong - this became part of the required reading for the Compute Governance Career Accelerator Fellowship.
- (Jul) Started Safe AI in Law, an organisation to build capability in the legal community
- Connected with people in AIS networks in London, Berlin, Cambridge, Oxford, Lisbon, Phillipines, and Australia
Fellowships & Courses
- (Apr) Blue Dot AGI Strategy course
- (Aug) GenStream Residency (12/400 accepted)
- (Sep) Blue Dot Frontier AI Governance course
- (Sep) SPAR Fellow (600/8000 accepted) Project: Bridging mechanistic interpretability and law (mentor: Elija Perrier)
- (Oct) High Impact Professionals Accelerator Program (<10% accepted)
- (Oct) CEA Bootcamp
- (Oct) Blue Dot AGI Strategy course (facilitator)
- (Oct) Compute Governance Career Accelerator Fellowship (lecturer)
Conferences & Events
- Oxford Technical AI Safety Conference
- EA Global: London
- EA Global: New York
- EAGxOxford
- ML4Good EU Frontier AI & Law Seminar
- Apart Research - digital minds (sprint)
- Apart Research - incident response (sprint)