In terms of how most people use it, it really does seem to rely on stronger models to infer and fill in their needs. So it’s understandable that even the most powerful W3 circle big shots are still stuck in how they understand the AI space.

For example, if you want to participate in a certain airdrop project, and you’re a big-name figure in the crypto world, you go research it yourself, and once you feel it’s workable, you use it there. Then you slowly iterate and continue studying it—many of these efforts involve systematically producing investment research reports for in-depth research and analysis. Why, once you come into the AI circle, does it turn into using it in an unsystematic way on your own? At that point, the gap widens again.