You asked ChatGPT to explain self-attention. It gave you a paragraph that sounded right. You nodded. Two hours later, you couldn't explain it back to yourself.
That's not a knowledge gap. That's a prompting gap. You asked the model to explain, but you never told it how deep to go — or at which level you actually needed the explanation. So it defaulted to a single-register answer that optimized for sounding helpful rather than for making you understand.
The Feynman Technique fixes this — not by asking the AI to "explain simply," but by forcing it to explain the same concept at four distinct cognitive levels, each one exposing a different layer of the idea. The result isn't one explanation. It's four passes over the same territory, each at a different altitude. And the gaps between those layers are where genuine understanding forms.
What the Feynman Technique Actually Is
Richard Feynman — Nobel laureate in physics, legendary teacher — had a learning method that was disarmingly straightforward. Pick a concept. Try to explain it as if teaching someone who knows nothing about it. When you hit a wall, go back to the source material. Simplify again. Repeat until the explanation is clean. As documented by Farnam Street, the power of this method lies in its brutality: it reveals exactly where your understanding is performative rather than structural.
