In the context of AI model capabilities, Ajeya Cotra uses the term “situational awareness” to refer to:

a cluster of skills including “being able to refer to and make predictions about yourself as distinct from the rest of the world,” “understanding the forces out in the world that shaped you and how the things that happen to you continue to be influenced by outside forces,” “understanding your position in the world relative to other actors who may have power over you,” “understanding how your actions can affect the outside world including other actors,” etc.

Alternatively, from an ML-perspective, situational awareness can be characterized as a strong form of out-of-context meta-learning applied to situationally-relevant statements.

“Situational awareness” of course has a broader meaning outside of the AI context. Even within the AI context, it’s used to refer to both “the awareness that AIs have about their situation” and “the awareness that relevant human decision-making bodies have about the AI situation”. Leopold Aschenbrenner’s Situational Awareness is an example of the latter.