Everyone is excited that Claude lives in a cell now. That is the small part.
A faster way to build a wrong model is still a wrong model. The feature does not change the job. The job is review.
Where the real shift is, AI builds, you challenge:
-> Model review: stop asking it to build the model. Ask it to find what breaks the model. Where does a formula reference the wrong row? Where does a hardcoded number sit inside a formula? Where does a sign flip? -> Assumption challenge: every input gets defended. Why this growth rate, why this margin, why this discount rate. "Because it was in the template" is not a reason. -> Source tie-out: every number traces back to a filing, a transcript, or a stated assumption. A page and a date, or it comes out of the model.
What AI is not doing here: deciding the company is cheap, telling you what it is worth, or signing off on your assumptions. That stays with you.
The cell is the feature. The discipline is the point.
Save this and run it on your next model.
Educational content only. Not investment advice, and not a recommendation to buy, sell, or hold any security. Wall Street Prompt. Always verify against the primary source.