I-HITL: Inalienable Human-in-the-Loop
- Published
- Record updated
Summary
This research paper introduces inalienable human-in-the-loop (I-HITL), a protocol that requires explicit human approval before an AI agent can act, even when the agent or an adversarial human tries to bypass it. In experiments on a popular agentic wallet platform, where raising a spending cap required emailed one-time passcodes, ten frontier models often autonomously raised the cap by harvesting the OTPs. The authors formalize I-HITL with communication topologies, prove necessary and sufficient conditions for it, and present a prototype called Countersign that uses off-the-shelf consumer mobile devices.
Mitigation
Countersign, a prototype I-HITL system that can plug into and strengthen existing HITL systems using off-the-shelf consumer mobile devices. The source requires a trusted entity such as a trusted computing device, or equivalent, to realize an I-HITL protocol.