Conference
Open access
Oct 2026
The Safety Exposure Paradox: How Frontier AI Safety Research May Amplify Cyber Risk in Financial Infrastructure
Frontier artificial intelligence laboratories now publish detailed accounts of how their models behave under adversarial pressure: red-team transcripts, capability thresholds, interpretability findings, and responsible-scaling frameworks. These disclosures are widely treated as safety infrastructure, built to keep dang...
Abena Primo, Azubike Okpalaeze, Al Amin et al.
· Proceedings of the 7th Natio... · 0 citations