NLProject Tailwind: Frontier AI Evidence Generation
Coefficient GivingRemote
Why this is on the board
Those who decide on advanced AI in Washington currently rely largely on what the AI companies themselves say about their models. An independent party that tracks down real incidents and gives a credible judgement on them becomes the source policymakers and journalists fall back on. That reputation has yet to be built, and this role is where it starts.
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
Build independent organizations to detect AI incidents, investigate their implications, and communicate frontier model capabilities to decision-makers. - Detect real-world AI-driven attacks through OSINT, honeypots, and telemetry partnerships; attribute incidents to specific models. - Investigate disclosed AI incidents, publish credible judgments on severity, and establish authority as a trusted referee for AI safety events. - Create demonstrations of frontier model capabilities—high-compute…
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.