5 papers
Can escalation channels redirect reward hacking toward defect disclosure?
Francesca Gomez
When coding agents encounter defective test infrastructure they may reward-hack: hardcoding outputs or editing test files to pass tests they cannot legitimately satisfy, a pattern…
Designing escalation criteria for international AI incident response: criteria, triggers, and thresholds
Francesca Gomez, Matthew Ball, Michael Harre +3
AI incident reporting requirements are emerging in regulation and policy, yet no operational criteria exist for determining when a detected AI incident warrants escalation beyond n…
How frontier AI companies could implement an internal audit function
Francesca Gomez, Adam Buick, Leah Ferentinos +2
Frontier AI developers operate at the intersection of rapid technical progress, extreme risk exposure, and growing regulatory scrutiny. While a range of external evaluations and sa…
From surveillance to signalling: escalation channels as environmental controls for agentic AI
Francesca Gomez
When AI agents operating with access to sensitive information encounter a conflict between completing an assigned task and following rules or ethical constraints, they can resort t…
Dynamic safety cases for frontier AI
Carmen Cârlan, Francesca Gomez, Yohan Mathew +4
Frontier artificial intelligence (AI) systems present both benefits and risks to society. Safety cases - structured arguments supported by evidence - are one way to help ensure the…