There appears to be a few bad actors in the Lemmy community spreading ragebait for personal gain
In fact, the same mechanisms added at the beginning to catch bad human written code are the exact same that catch bad AI code
- No, the Linux kernel DOES NOT use AI to automatically approve code for inclusion. As far as I can tell, there are no plans to do so
- The Kernel actually has a huge number of maintainers. As part of the process, all patches are reviewed by a human who provide feedback as part of the process, and patches get bounced back and forth multiple time.
- Patches don’t just magically go straight to master. The maintainers generally have their own branches where it goes to, then it goes to staging trees like Linux-Next. It goes through many eyes, and many tests by the end. The idea that AI code is overrunning the kernel is rediculous
- Maintainers also check that the code meets coding guidelines and is written properly, and solves the problem they are trying to solve. Yes, they also ensure the code isn’t slop
- LLM’s though might be used to identify obvious problems with the code (which they are VERY good at), or even identify code that doesn’t follow coding guidelines.
- The only thing that matters is the human who submitted the code, as they should be checking the code submitted anyway (including LLM generated code)
Writing bad code isn’t something new, and isn’t unique to LLM’s. Coders (and it seems some “journalists”) have been writing slop since computers were invented.
The process hasn’t gotten less stringent (probably even MORE stringent as there is an additional set of robotic eyes that can also check), and the same mechanisms that were put in place to catch badly written code 10 years ago, are still there now… With no changes.
True, but the seal’s broken now
There is truth to every bit of propaganda. AI is being used in a limited fashion on the Kernel. In the long term AI will likely decrease the contributors as already seen with AI blowing up some traditional communities. This is a very real problem and people should be concerned.
The sky is not falling. The future is very uncertain though and this could be the start of something very disruptive to the open source community.
AI will likely decrease the contributors as already seen with AI blowing up some traditional communities.
That’s an interesting problem that I haven’t heard anyone bring up before. I wonder how much of an impact this actually has.
There is truth to every bit of propaganda
This is not true.
Let me clarify, there is a bit of truth in every piece of propaganda. That is what makes it effective.
Still not true. People often just lie or make things up.
I agree with you that the Kernel isn’t doing slop. But there are a few points I want to address, maybe even disagree. First let me say that people have different definition of what slop means. That is why we don’t always agree on statements involving slop.
he Kernel actually has a huge number of maintainers.
True, but its still not enough to handle all slop. One problem is, its not always apparent at first glance. And more code is generated than ever before, surpassing the capabilities, time and nerves of all maintainers. That means they start using Ai to combat the amount. And that’s an issue on itself.
Writing bad code isn’t something new, and isn’t unique to LLM’s. Coders (and it seems some “journalists”) have been writing slop since computers were invented.
This might be true, but that is not a reason to accept or normalize slop. It does not matter how much slop was before, we always try to minimize it. The difference to before is, that nowadays its easier and faster to write (or generate) slop in mass that looks legit. Finding out its bad code or wrong can suck up huge amount of time of maintainers, which would have worked on something more productive instead.
The only thing that matters is the human who submitted the code
For you maybe, but there are issues using Ai regardless of how the code looks. And I don’t even talk about the amount of code it can generate and outpace every maintainer capacity. I mean stuff like how the LLM was trained on in example, and if the licensing will become an issue one day.
While the Linux Kernel operates very securely and in a controlled manner, that we cannot call it slop (yet). But there are inherent problems using Ai that are unsolved.
Writing bad code isn’t something new, and isn’t unique to LLM’s. Coders (and it seems some “journalists”) have been writing slop since computers were invented.
Yeah, this was the dumbest point OP made IMO, there are non-experienced people but they don’t become Kernel maintainers or contributors. He’s admitting LLMs slops this up and LLMs are yet Kernel maintainers and contribitors.
It does not matter how much slop was before, we always try to minimize it.
There is no policy that will stop that. It’s happening whether you want it to or not.
What every project needs is a way to handle submissions that ensures a level of quality. The kernel has this. Also needing to police whether code was llm generated would be an increased burden.
True, but its still not enough to handle all slop. The number of maintainers have also grown though dramatically… Also, don’t forget that LLM’s can also be used to identify bugs in patches more quickly too… So, more are getting submitted, but, lots of bad patches are being rejected faster too, with less effort (that required human intervention in the past).
This might be true, but that is not a reason to accept or normalize slop. Define Slop…
The difference to before is, that nowadays its easier and faster to write (or generate) slop in mass that looks legit. Finding out its bad code or wrong can suck up huge amount of time of maintainers, which would have worked on something more productive instead. We’ve all spent hours doing that though on our own code. Again, the buck stops with maintainers. One thing that Maintainers are going to do though, is get them to prove the code first and test the code in other trees first, before even trying to push to main
There’s always been code that has been bad, like Linus never liked the Con Kalivas performance patches, or Reiser4.
It is also finding a lot of security issues and bugs faster too though
I’ve actually written optimisations for code that I thought would be faster, and turned out to be a waste of time in production
For you maybe, but there are issues using Ai regardless of how the code looks. Both sets of code need the same level of examination by humans anyway. In fact, in some cases, the AI code I’ve seen has been better than the code i wrote
I mean stuff like how the LLM was trained on in example, and if the licensing will become an issue one day. That is true, and I agree with that. That’s the one point that does stand out, but doesn’t have anything to do with the claim by certain people in the community about the Linux Kernel being overrun with bad code.
What I would like to see is a LLM that’s developed exclusively using Open source compatibility licences with FULL traceability to sources uses, and a way to specify licencing allowed (or something similar).
There is one issue with AI that you didn’t mention however. The biggest problem with AI is that it adversely benefits closed source companies via Open source code. Because closed source code isn’t easily accessed by the AI for training, but Open source is. The other issue is that hackers won’t avoid LLM’s based on the ethical issues, so, unfortunately, it forces developers to use them to keep up with the cat and mouse game.
As a developer, am I happy about AI? For code review yes, but, it makes my job suck and not fun. And no, I’m not happy about the ethics… There are a lot of things that sh** me, like AI companies SHOULD be forced to pay royalties to everyone they stole info from.
However, what sucks, is that we can’t pretend it’s not a thing, and if Linux wants to remain competitive, it does need to use it.
The best thing we can do though, is petition to tax AI companies and pay producers of the information. That would make AI unprofitable immediately and kill the industry
Nobody ever remembers the environmental issues of using AI, like climate change and displacement of poor people. And that’s happening inside the Evil Empire itself, imagine what will do to the rainforest and other ecosystems.
That’s because it could be done in a way that doesn’t cause those harms to people or the environment. The software itself doesn’t dictate that.
It’s NOT being done in that manner, but that’s a capitalism problem, not inherent to the software. This is just a computer program we’re talking about here.
Everyone knows about that… But, that doesn’t have anything to do with “AI Slop”. that comes down to ethical factors and is unrelated to the quality of the code
Funny that someone thinks it’s acceptable to compartimentalize the main issue with AI usage while being trigger happy to label outcry as bad actors.
Is it normal for you to join discussions, and start talking about something different?
You’re certainly free to start a discussion about environmental usage, in which case I’ll agree
But it has nothing to do with the code being produced
But that’s like saying “I made a machine that kills puppies and plants trees, and every time I plant a tree people complain about dead puppies! I wish they could separate the impact of the dead puppies from the impact of planting trees.”
No. It’s like invading a thread talking about giving snacks to dogs. And telling everyone that there is a machine that kills puppies and turns them into sausages
Completely different.
Again, you’re free to start a thread on the environment. But it’s a complete deviation from this conversation
What I would like to see is a LLM that’s developed … with FULL traceability to sources uses,
That’s not really possible. I mean, you can list the training data, but you can’t attribute a specific output to a specific piece of training data, because all of the training data influences all of the weights. It’s not possible to know which training data influenced which weights and how, because that data is not retained (and it would be massive and difficult to interpret if it was -on the order of weights × number of training context windows, and the weights alone are already terabytes)
Because closed source code isn’t easily accessed by the AI for training, but Open source i
Besides all those Ai users, where code is uploaded to their servers.
On commercial accounts, you can turn off training, and they have policies generally to avoid code leaks (another bad thing for open source, the free accounts you generally can’t, because you are the product).
Greg KH (the second in command of the Linux kernel, below Linus) in a recent talk said anything uploaded to a corporate AI is going to be used for training, and to assume anything uploaded or pasted into it is insecure and certainly not private.
If you believe that they’re not using corporate code to train their models… These guys are all total lying sociopaths who think they’re on a mission from the machine god, everything is permissible in their minds.
even identify code that doesn’t follow coding guidelines.
Code linters have existed for decades, that doesn’t require LLMs
That’s how I feel about autocompletion and scaffolding, we’ve had LSPs and code generators for years, and they’re exponentially less wasteful
AI code review finds a lot more bugs (such as logic bugs). But yes linters can be used for that specific case
Not all guidelines are enforceable with linters. For example one of our code bases at work defines when to avoid mock tests, how integration tests should look like, how a module’s architecture is supposed to be, how the naming convention for components of the different layers are, etc.
Those are super ugly to formalize into a linter but super easy to check with an LLM.
Huh. Really? How is the LLM checking it? Is it writing code to check those specs, or are you just along it directly if the code matches the specs?
Normal text based analysis; which LLMs are pretty good at. So our coding agents can follow the rules during development and the review agent can follow it during review. Only static code checks are enforced in the CI pipeline; the review agent only leaves remarks. If we decide to not give a damn, we can still merge it. But its hit-rate is relatively high; so typically when it finds something, it’s valid and its recommendations for fixes are also often pretty good. This helps already iterating on a PR on your own before you pull human reviewers in.
There appears to be a few bad actors in the Lemmy community spreading ragebait for personal gain
There’s no reason to attribute it all—or even most—to malice or opportunism.
Why do I see all comments to this thread with vote = 0?
I’ve seen some rage bait and people spreading falsehoods about Linus and his outlook on LLMs… but what personal gain is there to be had? I would’ve assumed all the misinformation I’ve seen is just doomerist outrage.
what personal gain is there to be had?
The satisfaction of being “on the good side” and “sticking it to the man” with righteous outrage, as usual with these things. I fucking hate AI, but trying to tear down Linus (and the rest of the Kernel team) over them trying to see if it makes sense to use is silly. They aren’t planning to drop their rigorous QA or to accept code the submitter can’t explain/maintain. Ethics are a concern, but since the end result will be FOSS I personally don’t care as much either (YMMV).
I’m also really interested in hearing how this will go for them and what policies they will ultimately implement. I expect them to do this properly, which then means that everyone can point to their implementation when confronted with vibe coding at work. And then the suits will decide that they’d rather just do it cheap & fast, fucking everything up yet again, but at least they will have been told so.
I appreciate this level headed response. I tend to agree with you all around, and I think that no other humans on earth are more equipped to manage the use of AI than the team of mostly altruistic people who brought us OS choice. I find it hard to believe they will completely botch it or learn no lessons. I too dislike AI but I also recognize it may have some uses. Thinking it’s all bad, imo, is binary thinking. I think AI may end up being a net negative for humanity honestly, but 1) we don’t know that yet 2) the Assholes capable of regulating it dgaf. Regulation is really the only tool we have. Blind rage is not a tool in any scenario, including this one.
How many of you commenting here have skin in the game?
For most of us, we live in a tool-using culture whose digital infrastructure has become completely reliant on server-side linux, so its reliability (or lack thereof) can have real effects on our everyday lives. Can’t speak for everyone though
Do you know that Linux kernel has hidden DRM in it? Total betrayal, don’t know about you guys, but I’m no longer going to use it.
Source?
It certainly has the Direct Rendering Manager, which has no relation to any content control mechanism.
I’m not a participant in Linux development, are you a kernel developer/change reviewer?
