Debian is polling its developers on whether to permit or ban "AI" contributions.

[pndc@social.treehouse.systems]

@jackpearse @lproven That one's easy: people who submit AI code just can't resist boasting about the fact.

[sassinake@mastodon.social]

@lproven

One day we will wake up and NOTHING will work anymore.

NOTHING.

but all the poor countries - the 'shithole' countries? - yeah, they will just continue to bring and sell and buy food at the local market, and ride their bikes back to their casa, and cook on their woodstove, after getting water from the well.

but we will all die within a week.

[jackpearse@nrw.social]

@lproven

What I hate about this movement is: no one explains how you detect it? What about false positives?

I once contributed to OSS. My pull request was rejected. Because it had the „feeling“ of LLM. That feeling was me summarizing all changes of mine and a file listing. I wanted to safe the maintainers time. And exactly this „service“ was treated as „feeling“ like LLM made.

This is ridiculous. You work for days. And then people fuck you because they have a „feeling“.

[angelascholder@mastodon.energy]

@lproven BAN!

[jackpearse@nrw.social]

@pelle @lproven

„trust“ is no answer to this question. An LLM ban can only wirk if there is first a presumption of innocence of some kind. And, above all, there must be criteria. These must be formulated in abstract terms. Otherwise, it becomes arbitrary and a matter of personal belief.

Once a hand made pull request of mine was rejected because of a „feeling“. This is fucked up.

A ban also needs abstract criteria.
Otherwise contributions will die out.

[sunweaver@fosstodon.org]

@RueNahcMohr @lproven Send me an email at point me to your code projects. I can help bringing those to Debian if quality standards and license/copyright attributions are sufficient. If not, I can advise how to fine-polish. Mail to sunweaver at debian.org

[jackpearse@nrw.social]

@pelle @lproven

Again. I am not asking about the BAN. I am asking about the METHOD.

HOW do you detect Slop?

This is the most important question. And if the answer is „by a feeling“. Then we have lost science by far and entered the realm if esoteric.

[jackpearse@nrw.social]

@Cal @pelle @lproven

No. People have already set the boundary and excluded people who write good code. And accuse them to have used LLM „by a feeling“. And this starts to happen more often. And I can accept a ban only if there is a method to detect this.

„A feeling“ is no computer science.

So, how is „malware“ detected, please? What is the formal method?

.

[ferrix@mastodon.online]

@lproven @joe if we lose Debian what is left

[jackpearse@nrw.social]

@lproven

Thorvald says LLM is allowed to be used for the kernel. Then LLM use is banned by Debian.

Is the Kernel then banned?

In biology and medicine they use AI to detect cancer cells. Is this also banned then? Or are these type of programs allowed then?

What is with Computer Scientists developing free LLM tools. Also banned?

This kind of approach isn‘t realistic.

The overall issue is a quality problem. And needs tools to maintain quality. A ban is simply symbolic politics.

[jackpearse@nrw.social]

@pndc @lproven

I also assume the ban will happen. But I give that ban one year. Then it will be released again. Because the ban will not solve the quality problems.

The true problem is that tests have no real
Meaning anymore, if LLMs can modify them. Tests coild to be separate repositories, which cannot be touched vy everyone. Or we could invent tools which check the grade of code modifications?

[jackpearse@nrw.social]

@pndc @lproven

A tool like a linter which detects LLM code smells could be introduced. Which checks if SOLID and DRY was violated badly. This can never be perfect. But why can’t we find a way to automatically scan code upfront with abstract constraints.

Even humans can write really ugly code. Gosh, how often have I seen ugly code of beginners.

There must be a scientific way of maintaining the quality. Instead of „politics“ and „feelings“.

[cainmark@mstdn.social]

@lproven

BAN.

[lproven@social.vivaldi.net]

@ferrix @joe Apart from Alpine, Arch, Chimera Linux, Mageia, OpenMandriva, Slackware, PCLinuxOS, and dozens of others not based on Debian, plus the SUSE and Red Hat families, you mean?

[hillu@infosec.exchange]

@jackpearse You may find that reading the different proposals up for vote will help in answering your questions @lproven

[ahltorp@mastodon.nu]

@jackpearse @pelle @lproven A ban will:

1. Make people stop bragging about their use of ”AI” in connection with the project.
2. Make people unable to blame their faulty contributions on ”AI”.
3. Make people unable to blame ripped off code on ”AI”.
4. Make incontrovertible evidence of ”AI” use grounds for banning them.

[pelle@veganism.social]

@jackpearse
> Once a hand made pull request of mine was rejected because of a „feeling“.

that's really sad.

a great anti-AI cartoonist (who posts on fedi) is also being labelled by instagram as »likely to be AI«.

yes, you have to assume the best from others!

but that is made difficult, because those who share AI slop often deliberatly don't disclose it (whether code, images, websites or videos, ...) making everyone suspicious.

perhaps doing a few openly anti-genAI posts on masto should help others know where you're at?

if we compare again with veganism: it would be difficult to be vegan if every kitchen deliberately lied about putting animals into food! but, also, vegans either eat at venues that are clearly labelled vegan — or they ask about the ingredients three-four times to make sure.

[longplay_games@mastodon.gamedev.place]

@lproven @ferrix @joe RedHat (or IBM, however you wish to look at it) is so incredibly all-in on LLMs it's not even funny.
https://www.redhat.com/en/about/press-releases/red-hat-launches-llm-d-community-powering-distributed-gen-ai-inference-scale
https://www.redhat.com/en/blog/accelerating-open-source-development-ai
"Red Hat believes that AI offers tremendous opportunities for open source projects and contributors. "

[longplay_games@mastodon.gamedev.place]

@RueNahcMohr @lproven The trouble with LLM code in a project like debian is simple, deep, and obvious.

If someone is using an LLM instead of learning, providing official feedback to the LLM instead of growing, and providing the LLM's reply to the feedback instead of engaging...

You're not training the next generation of developers. You're training a model.

[ruenahcmohr@infosec.exchange]

@Longplay_Games @lproven

oh there are tonnes of issues, mostly that AI code is being accepted into the kernel debian is using. I dont know if thats better or worse than all the sabature coders who have been pushing exploits and garbage into the OSS community for over 10 years.