Cognitive Support for Rebel Agents

Rebel agents are intelligent agents that can oppose goals or plans assigned to them, or the general attitudes and/or behavior of other agents. They can serve purposes such as ethics, safety, and task execution correctness; enhance social co-creativity, and provide or support diverse points of view. We previously proposed a framework for AI rebellion and explored various scenarios that have positive AI rebels as protagonists. In recognition of the fact that, in human psychology, noncompliance has profound socio-cognitive implications, we now explore several mechanisms that could further support rebellion in cognitively-complex AI agents. We focus specifically on social awareness and counternarrative intelligence: a new term we propose that refers to an agent’s ability to produce and use alternative narratives that support, express, and/or justify rebellion, either sincerely or deceptively.