A present day AI makes an educated guess which existing source code snippets best match the request, does some testing, and submits code that it judges is most likely to pass code review.

That's still on the human that opened the PR without doing the slightest effort of testing the AI changes though.

I agree there should be a lot of caution overall, I just think that the problem is a bit mischaracterized. The problem is the newfound ability to spam PRs that look legit but are actually crap, but the root here is humans doing this for Github rep or whatever, not AI inherently making codebases vulnerable. There need to be ways to detect such users that repeatedly do zero effort contributions like that and ban them.

[–] unwarlikeExtortion@lemmy.ml 1 points 7 hours ago (1 children)

That's still on the human that opened the PR without doing the slightest effort of testing the AI changes though.

That makes sense when talking about people's accounts.

A "Claude" account serves PR (as in public relations) purposes, and having to do a stringent human review before submitting a pull request is bad for PR.

Which by no means is me saying submissions from the Claude account need to be banned, but that the "Claude" account's goals are probably to have Claude do all of this "himself" - which is a recipe for disaster.

[–] iByteABit@lemmy.ml 1 points 2 hours ago

I think Claude account PRs should absolutely be banned, that's the easiest counter measure to implement too

[–] BreakerSwitch@lemmy.world 6 points 1 day ago

Yes, it is their fault, and also, that fault is a widespread problem