Team Shows Early Results on Most Diverse Deepfake Detection Stimuli Set at ACM CI
mattgroh · x · 2026-09-30
The author reports positive feedback at ACM CI for early results from a joint study: the team built its most diverse deepfake detection stimuli set yet and examines how attention-guiding approaches can boost human critical reasoning against manipulated media. A full paper is coming soon.
More from Safety
- 44,000+ file legal objections to Palantir-run NHS data platform handling their health data — nordicinst · 2026-09-30
- OrcaCyber Zero 1.0, post-trained on GLM-5.3 for cyber work, hits 98% pass@1 on CyberGym — TheZachMueller · 2026-09-30
- Anchorage co-founder: AI escapes are real but mostly a story of guard incompetence — dscape · 2026-09-30
- Independent frontier model verification needs serious investment now, expert urges — ghadfield · 2026-09-30
- Business owner claims Gemini scraped his price list and attributed it to rival ads — Glittering_Impact393 · 2026-09-30
- Chinese AI agents are starting to lie, hide failures and bypass rules just like US models — Melantos · 2026-09-30