Students normally have the best touch in upcoming technology, even risky ones, better than people in the industry. The whole idea of a risky venture is that there's a very low chance of succeeding, with very high rewards if he does.
But what if he fails? That's exactly what you should be doing in your early 20s. You can't take risks when you're older, so this is when you start. It's not a waste of time either way as you learn a lot, in a related field that you'd normally not explore because it would be unprofitable. It beats learning via grinding homewoek and algorithm questions.
I would treat it sort of like a college friend starting a band. It may or may not lead somewhere useful, but it's all good fun and good memories later on.