The problem is as usual, users. AI for maintenance (updating & testing deps, re-writing parts to run with next major version of lib) is pretty low error % and just need some supervision and can make it more effective, in hands of people already familiar with project

But the flipside is of course users that are clueless won't now be stopped by "can't make a PR", they will throw prompt at AI and send it when the AI decides it's good enough