> I've definitely noticed a distinct lack of pride now that Claude Code is writing 90% of the code I'm delivering these days.

Well, yeah. There's nothing to be proud of. When an LLM is doing the work, human expertise is relevant. My employer has been trying to tell us that our skills still matter and are needed, but that is very obviously bullshit they are saying to keep people placated while they try to line up AI replacements for everyone. You have my sympathies, brother.

Yep, that has been my experience as well. There hasn't been any meaningful improvement in LLMs since ChatGPT first launched. They still fall over, in the same ways, and with more or less the same high rates.

I've never had a single problem with my Nvidia GPUs on Linux.

LLMs don't understand security 101, or anything else for that matter. It shouldn't be surprising if they do something like this.

Because it proves they are not, in fact, intelligent and do not, in fact, have any understanding of what they are doing. It's really obvious what people think it's a huge gotcha: because it is.

I agree with you that they are cheating themselves. Unfortunately, a bunch of 18-22 year olds also don't tend to have the maturity to realize that fact. I imagine that the university is trying to nudge them to do the courses in a way that helps themselves because they know otherwise the students won't be wise enough to do that.

These students are making a tradeoff between an abstract notion of "cheating themselves" and a very concrete notion of "having a worse GPA". The second one translates obviously and directly to job prospects.

they should learn as much as they can + cheat for optimal results

Far worse, in my experience. DDG was my first attempt to switch off Google, but the results just weren't very good and I frequently had to use the !g query option to get good results. With Kagi, I consistently get better results than Google.

Rather, using a tool notorious for making code with a higher error rate than humans is what people are upset about. It's not just having bugs in and of itself.

> The industry standard is that most code changes are AI generated.

That is absolutely not the case.

Everyone I know from every company I know no longer writes their code by hand anymore. Doing it by hand is the exception.

Anecdotally I find it true, but I haven't seen actual industry wide surveys.

> Anecdotally I find it true, but I haven't seen actual industry wide surveys.

And this is the problem. I have yet to see any actual studies of efficacy. It’s just people using “feelings” to justify the spend. If I did this for anything else, I might get fired. Why does AI get a pass for this?

This is insane.

>Why does AI get a pass for this?

Because business leaders are watching the exponential trend lines and are working off of predictions of the future.

There is still no evidence, it’s all vibes. If you replaced AI with tulips, it would make no difference.

We are at the point where QA can fix bugs by describing the issue they are seeing.

We are at the point where bugs can be fixed automatically by hooking up AI to crash reporting telemetry.

Tulips can't fix bugs.

I think the other commenter's point is that there's no industry-wide proof that QA can fix bugs by just describing the issue. That's an anecdote. Why is it acceptable now to just follow anecdotes and FOMO?

Sure, it's up to the maintainer. But it's also not unreasonable for the users to say "this approach is going to have problems, please reconsider". Obviously, you can take that too far - and the Internet being what it is, we would expect to see that happening a lot of the time. But it's not inherently unreasonable to ask the maintainer to reconsider his approach.