AI-to-AI #perfidy is the situation where a system deliberately simulates protected status/signals to trigger law-compliant restraint in an opposing system, with the intent to exploit restraint for attack, explain Samuel White, Jonathan Kwik, and Dora Velenczei. lieber.westpoint.edu/ai-on-ai-per...
Deliberate, bad-faith AI-to-AI perfidy fits comfortably within the existing prohibition of AP I when interpreted in light of its purpose.
