Hacker News Reader

Comment by qsera | Hacker News Reader

>Apple, Mozillia, and Firefox just released a record number of bug fixes because of AI.

A hidden metric here is the number of new bugs created by these fixes. If an LLM creates 10x bugs but create 5x more new bugs, then it is not really an improvement. Because that is 5x more bugs that the user has to observe and report (unless it is a security vulnerability, which the LLMs might detect) before it can be fixed.

View on HN

Also there's the "let's make time to work on those bugs by using this new tool". Time that wouldn't have been allocated otherwise.

...allocated on not fixing the bugs, sure. and AI does not decide what you spend your time on.

The LLMs find the bugs and the users will get exploited unless fixed. Sure am glad the choose fix them.

This is effectively a claim that Mozillia, Apple, Google's developers are idiots and irresponsible and just accepted the fixes without checking them. That's a pretty strong claim which I think you'd have a hard time proving or even finding any evidence for

I mean no such claim. One does not have to be an idiot or irresponsible to overlook an unintended consequence of some code change in a very large, legacy code base.

If that claim were even reasonable in the first place you would have to accept that they were idiots long before by allowing bugs to be introduced in the first place.

The presence or introduction of bugs in software does not mean it’s developers are ‘idiots’ unless practically everyone who has ever written code is an ‘idiot’

Yes I didn't say this. I'm using it to show how absurd the original claim was.

I think LLMs are particularly effective at turning once-competent developers into idiots. (“Just ask Claude.”)

> A hidden metric here is the number of new bugs created by these fixes.

If the project has proper test coverage, there should be no significant number of new bugs. This is no different than the possible regressions added by a human-implemented feature.

And LLMs have made implementing a massive number of tests far easier and faster than before the LLM era.

Reply on news.ycombinator.com