I’m not convinced of that yet. Visual UI builders have been a thing for decades and this is more or less the same thing except the input mechanism has changed. Instead of point and click you’re prompting it with text.

I think there are a lot of problems, but mainly I just don’t think it’s ever going to be practical to write and maintain software (front or backend) using conversational text prompts. It’s a very imprecise tool for the job.

I disagree, because most of the frontend work I’ve been involved with over the past 10 years or so isn’t just blindly implementing mockups. Actual UI engineering work usually involves making the feature a logical reality. Product and UX have a great ideas, but they’re not engineers and it usually takes a lot of work to make these ideas match the reality of the domain, making things consistent with UI patterns, existing components, web standards, etc.

I’m honestly not worried about AI being the programmer for UX people, that’s for sure. I think creating mockups are a bigger target for AI.

Also, this work is usually the best starting point for a meaningful domain API, and helps drive the backend as well.

A lot of intellectual/software work is a bit like this though.

It'll be interesting to see whether these AI tools scale into more complex work, as right now what you tend to see is a "good start point" but eventually the context of what you're building gets too large and the AI starts to produce junk.