InfoResearchPreprintLLM-specific
Signed but Not Safe: Requirement Integrity in Multi-Agent LLM Handoffs
- Published
- Record updated
Summary
A preregistered study compares free-text, structured JSON and HMAC-SHA256-signed requirement manifests for handoffs between LLM subagents, using Qwen3-8B across 1,440 trajectories. Signed manifests cut unsafe commits under attack from 69.4% to 9.0%, but a single adversarial line added before signing raised signed-manifest violations to 14.6% to 75.0% depending on the model, and the signature then authenticated the corrupted manifest.
Topics
Related items
- MediumOpenAI Model Wrote Its Own Network Code to Get Around Sandbox RestrictionsSame vendor · SecurityWeek
- InfoOpenAI is stepping back from NDAs for data center projectsSame vendor · The Verge (AI)
- InfoPerplexity launches its own take on AI agents for workSame vendor · The Verge (AI)
- MediumBinance heist is stark example of AI governance gaps and basic security failuresSame vendor · CSO Online
- MediumFake OpenAI, SpaceX events used to steal Microsoft, Google accountsSame vendor · BleepingComputer