every single published study shows that AI makes developers worse.

Can’t thank you enough for this list!!! I asked anti AI folks and the best I got was a study on AI assisted essay writing with 8 people per control group. The rest were insults, mockery, shifting goalposts, so naturally, as someone who just arrived at the fence, I found myself siding with the less troll-ey side. My guess is this is more an issue of “worse than an enemy is an incompetent ally” because any single study in your list would have done it for me.

This study has some positives on AI use in coding, but I’d exercise caution due to the tabloid-ey nature of using the term “AI-slop” everywhere in a study that’s supposed to be unbiased.

page 9’s wording does state that it is maintainer-estimated, and specifies the terms “low-quality” and “slop” as the qualifier for the 25% figure

I’m not seeing it. This is the sole relevant statement on pg 9 where they simply define AI slop in a general way that avoids tying it in with their measurements:

“AI slop” are low quality AI-generated contributions that are unlikely to be worth the review time they demand.

In addition to the explicit ‘and/or general AI use’ statement, there are several red flags here that make me think this is a low quality study to be taken with salt. Or maybe I’m cynical and using the term ‘slop’ to push a research study is both unusual and suspect.

you’ll see that “slop” in the survey does not refer to AI-driven PRs in general, but to specifically low-quality AI-generated PRs.

This part is actually unclear. The question posed to survey participants that allowed researchers to determine a PR was “AI slop” is left out of the report. And this line under the chart showing “25% slop etc.” is troubling:

Average estimated percentage of incoming PRs that are low-quality and/or AI-generated

How this particular question was framed is of critical importance here and without it, it’s easy to assume that all reported AI use was broadly categorized as “slop”

Sadly I could not find a way to access the full list of questions used in the survey to confirm this.

It looks like it was more of a survey and afaict, participants weren’t given an opportunity to distinguish between “AI slop” and high quality AI assisted contributions (so were all AI contributions reported as “slop”?!)

To answer your other questions, this is the relevant part:

Across the board, genAI reduces estimated task time across all six measured activities in the pooled sample, with the largest gains in code authoring (−52%) and code review & feedback (−44%), and smaller but still meaningful reductions in technical discussion (−20%) and discussion on proposals (−19%)

Productivity gains are most often reinvested in addressing technical debt [66%], adding new features [63%], and learning/upskilling [58%] … genAI productivity is invested right back into OSS projects. In other words, genAI is not shrinking the overall OSS workload so much as reallocating it toward highervalue and longer-deferred work, from debt paydown to skill-building.

It’s not all clear weather and blue skies though. My interpretation of the findings is that there definitely are challenges to address, but also that those that take the effort to address and integrate AI usage see the most in terms of net positives.

You can download the full report here.

What would be the point of jumping through these hoops in the first place? As OP pointed out, the disclosure requirement is itself irrelevant due to the amount of scrutiny the code goes through. So we’re forcing disclosure to… shame AI users? Discourage its use?

As time goes by, you just learn to understand that some people are just not worth your time.

No u

Children don’t typically express honesty with ‘fuckstick’. ‘Childish’ here is used to refer to the absence of capacity or ability to engage in constructive discussion. But feel free to keep painting ‘fuckstick’ in any shade you like. I personally prefer ‘Yo mama’ jokes when I lack the ability to win a particular argument.

Sure, but if ‘fuckstick’ is your sole ‘argument’, you can’t exactly call foul when someone declines to engage with you for being childish.

Do you have an actual counter to my original argument or are you just going to throw vague insults at anyone who disagrees with you like everywhere else you’ve commented in this post?

Sure, but calling people fuckstick isn’t used on a large scale to shut down conversation, there is a qualitative difference here.

Maybe if ‘fuckstick’ was followed by a coherent argument. But instead of providing one, OP just continues to throw insults about their reasonable perspective somehow being ‘smug’ in some way. I’d call that childish too.

90% of AI bros hide it and refuse to disclose and when they do are often extremely cagey about it

source?

Why is it that every single time you see someone trying to claim the role of ‘adult in the room’, they appear to be speaking from a perspective of being perfectly oblivious (albeit sometimes for the purpose of obfuscation)?

Did you miss the part where OP’s entire argument was ‘fuckstick’?

Yes, fuckstick. That’s the whole point to a lot of people. I swear, the level of smugness of AI true believers is totally inappropriate.

Because rules have to be enforced, and investing time and effort into detecting and punishing rule-breakers would be a pointless endeavor when, disclosure/AI or not, all code has to be reviewed anyway.

I have to say the same thing I said to the other commenter; I’m really not sure that “some people lie so nobody should care about the truth” is the argument you think it is.

Nice attempt at spin but relies too much on readers having room temperature IQ to believe that’s what my argument is. And just to be clear, by ‘Nice attempt’ I’m being sarcastic.

On the contrary, a small piece of text that says “this code deserves more review because humans may not have seen it” is absolutely necessary.

All this is doing is making it so I can hide that I didn’t touch the code.

Why are you assuming that everyone would honestly disclose AI use if it was a rule?

The honest ones that would disclose it to say “this needs more eyes, guys” can still do just that. The dishonest ones that would hide AI use will do exactly that whether it’s a rule or not.

The only difference here is there’s no punitive action to follow for non-disclosure because that would be a gigantic waste of time when, AI or not, all code needs to be reviewed anyway.

Even considering rhetorical goals and context, without “I particularly dislike this guy”, it’s an enormous leap to twist those words that way.

Stop blaming the pandas.

Thanks for your contribution to the pile.

One example.

Got something that has more than 18 participants split into 3 groups for a total sample size of 6 people per control group? Not to mention its focus on AI assisted essay writing, measuring cognitive debt (not decline), using EEG to infer performance of higher cognitive function, etc.

Vague criticisms, fallacious reasoning, highly emotive language, little response to attempts at discussion (though some might engage in disingenuous discussions of shifting goalposts here and there to discourage particularly vocal engagees from future participation). So I’d say that Linus is just the latest convenient scapegoat in an Anti-AI ‘trend’ that will have little impact on the rate of AI adoption in the long run.

Sucks that I waited until now to try to form a measured opinion on AI because now all the discussions on the internet are hot garbage.

As someone who did study computer science and has worked in embedded processor space

Using an argument from authority to denounce someone in authority and their arguments? Cool. Might’ve worked if I hadn’t seen you mention recently that we shouldn’t use AI ‘because Trump and his cronies do’.