Pick the Wave

Altman’s Idea Generation essay locates the largest returns in timing, not execution: spot a genuine tectonic shift and get there early. His practical test for a good idea is a specific feeling of inevitability, the ability to say, “I am sure this is going to happen, I’m just not sure if we’ll be the ones to do it.” He suggests studying what became newly possible this year versus last, since that delta points at the wave.

His canonical case is his own. Going all-in on scaling neural networks in 2015, years before it was consensus, was a bet on a wave he was sure would break, whoever rode it. The conviction was about the shift. It was not about OpenAI specifically.

Doesn’t transfer because timing is invisible in the moment. Wave-picking only looks smart in retrospect. The same “early to a shift” bet that reads as genius when it works reads as crankery when the wave doesn’t break, and being early and being wrong are indistinguishable in real time. Nobody rings a bell.