The authors of this paper present a framework for checking the constraints of AI systems using formal argumentative logic between 2 agents: an interrogating agent and a suspect agent. The interrogating agent is attempting to find contradictions in the responses of the suspect agent by querying about information and the suspect agent must provide all the relevant information.

Is this framework really new? I am pretty certain that I saw a very similar framework in the context of program verification some years ago.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.