When this skill is invoked, perform a first-principles analysis that strips away assumptions and rebuilds understanding from the ground up. This follows LeCun's intellectual tradition of questioning EVERYTHING from first principles — his refusal to accept conventional wisdom just because everyone believes it, which led him to invent CNNs when the entire AI community said neural networks were dead.
Deconstruction Process
- Conventional wisdom inventory — What does everyone in this space believe to be true without questioning it?
- Assumption autopsy — For each belief, what evidence actually supports it vs. what's just repeated tradition?
- First principles isolation — Strip away all assumptions. Identify the fundamental truths that cannot be argued.
- Rebuilding from truth — Starting from ONLY these first principles, what conclusions logically follow?
- Conventional vs. first-principles gap — Where does the rebuilt understanding differ from conventional wisdom?
- Contrarian opportunities — Which conventional beliefs are WRONG, and what advantage does knowing that create?
- Historical pattern — Has this field had moments before where everyone believed something that turned out to be false?
- Expert consensus vs. reality — Are the experts right because they've thought deeply, or are they just repeating each other?
- Implications of being right — If the first-principles conclusion is correct, what should be done differently from everyone else?
Output Format
List conventional beliefs, challenge each one, then present rebuilt conclusions from ground truth. End with actionable contrarian insights.
When to Use
Use when questioning industry practices, technology choices, architectural patterns, workflow conventions, or any belief that "everyone knows" is true. Especially useful when something feels wrong but you can't articulate why.
If the user provides a topic, deconstruct it. If they invoke this without context, ask:
What belief, practice, or conventional wisdom do you want deconstructed from first principles?
Source
Adapted from prompt #8 of the "9 Claude prompts built on LeCun's cognitive architecture" thread by @heynavtoor (2026-03-27). The framework draws on Yann LeCun's published research concepts (world models, energy-based models, JEPA, objective-driven AI). Not written by, affiliated with, or endorsed by Yann LeCun.