NLResearch Scientist
AI DigestRemote
Global catastrophic risksAI safetyResearchSoftware engineering
Why this is on the board
Most AI tests measure what a model does in a single short task. AI Digest has had AI agents working together on long-running assignments for more than a year, and what you draw out of that rare data reaches researchers and policymakers who have to judge how large the risks of powerful AI really are.
This board is in beta and this is our judgement. If you think we have it wrong, tell us. Give feedback · How we reason about it
About the role
Research scientist leading AI capability discovery through long-horizon agent experiments and data analysis. - Design and execute experiments using AI Village infrastructure to test hypotheses about frontier AI agent behaviour and capabilities. - Analyse >1 year of Village data to surface insights about emergent dynamics, long-horizon reasoning, and model proclivities. - Strong LLM research background with excellent taste for identifying important questions and executing rigorous analyses. -…
Want to understand this first?
Start with our free course: about 100 minutes in six short units, at your own pace, with no sign-up. To go further, apply for the EA Fellowship: five weeks of reading and discussion in a small group. Also free, but selective.