I worry that almost all the 2025 startups I've seen are AI app builders. Where are the novel new applications? I get that codegen is currently one area where AI does well, but it also feels like we're struggling with other use cases.

I’ve spent an enormous amount of time with practically every AI model out there, from coding to image-gen to video-gen

The tech is still simply too hard to use effectively for the vast majority of lay people, especially for anything beyond a cool product demo

Some of it is due to quality of the models, some of it due t quality of the tooling

Prompt engineering is still a skill and that’s beyond what a casual user can figure out

My optimism says the good new stuff is coming slowly because people who care about their craft and taking things slowly aren’t in any rush to get to market.