Longterm Wiki

Evaluating Model Attack Selection and Offensive Cyber Horizons

$41K
Funder
Recipient
Sean Peters
Program
Date
Dec 2025
Source
Notes

[Technical AI safety] Measuring attack selection as an emergent capability, and extending offensive cyber time horizons to newer models and benchmarks

Other Grants by Manifund

376
Showing 10 of 376 grants
Evaluating Model Attack Selection and Offensive Cyber Horizons | Grants | Longterm Wiki