Edit: ok so this whole text concoction is just a product placement for their own service GolemUI
Both can be "100% code written by AI" but there's a world of difference between the two.
In my experience, the more constrained the LLM is, either by existing patterns in the code base or by prior art in general, the better the LLM is. When I've asked it do architectural or design work, it has been much less successful. That's one dimension on which LLM capabilities vary. There are certainly more!
For vague, divergent tasks, the overall experience hasn’t been quite as good.