You ask an AI not just to answer your question, but to show its reasoning—to tell you why it arrived at that answer and what assumptions it made along the way. This exposes weak spots in the logic or reveals when the model is confident about something it shouldn't be, making you a more critical consumer of its output.
Metacognitive prompting involves asking AI to evaluate, critique, or explain the reasoning behind its own response — for example, prompting it to identify weaknesses in the answer it just gave or rate its own confidence level.
This self-reflective layer surfaces blind spots, improves output quality without requiring expert review, and builds a habit of critical engagement that makes every AI interaction more reliable.
Peri can explain this concept, give practical examples, help you decide whether it applies to your situation, or recommend a journey if appropriate.
Explore related journeys or tell Peri what you're working through.