From a technical perspective, you're absolutely correct.
From a regulatory perspective, it seems unlikely that most courts would appreciate the difference. In their mind - you run a website, and that website contains copyrighted content. Take it down.
You'd probably have to just blacklist the link in question to avoid a legal headache.
I've been finding it fascinating to see the different approaches to this by the big players. I would've definitely expected Wikipedia to have a blanket ban (even if hard to enforce), but it seems like they are allowed as long as you're not just generating the full article
Human oversight: The use of AI must always remain under human control. Its functioning and outputs must be consistently and critically assessed and validated by a human.
Quite. One would hope, though, that it would be clear to prestigious scientific research organizations in particular, just like everything else related to source criticism and proper academic conduct.
Sure, but the way you maintain this standard is by codifying rules that are distinct from the "lower" practices you find elsewhere.
In other words, because of the huge DOGE clusterfuck demonstrated how horrible practices people will actually enact, you need to put this into the principles.
Oddly enough nowadays CERN is very much like a big corpo, yes they do science, but there is a huge overhead of corpo-like people who running CERN as an enterprise that should bring "income".
Someone's inputs is someone else's outputs, I don't think you have spotted an interesting gap. Certainly just looking at the dials will do for monitoring functioning, but falls well short of validating the system performance.
The real interesting thing is how does that principle interplay with their pillars and goals i.e. if the goal is to "optimize workflow and resource usage" then having a human in the loop at all points might limit or fully erode this ambition. Obviously it not that black and white, certain tasks could be fully autonomous where others require human validation and you could be net positive - but - this challenge is not exclusive to CERN that's for sure.
It's still just a platitude. Being somewhat critical is still giving some implicit trust. If you didn't give it any trust at all, you wouldn't use it at all! So they endorse trusting it is my read, exactly the opposite of what they appear to say!
It's funny how many official policies leave me thinking that it's a corporate cover-your-ass policy and if they really meant it they would have found a much stronger and plainer way to say it
"You can use AI but you are responsible for and must validate its output" is a completely reasonable and coherent policy. I'm sure they stated exactly what they intended to.
If you have a program that looks at CCTV footage and IDs animals that go by.. is a human supposed to validate every single output? How about if it's thousands of hours of footage?
I think parent comment is right. It's just a platitude for administrators to cover their backs and it doesn't hold to actual usecases
I don't see it so bleakly. Using your analogy, it would simply mean that if the program underperforms compared to humans and starts making a large amount of errors, the human who set up the pipeline will be held accountable. If the program is responsible for a critical task (ie the animal will be shot depending on the classification) then yes, a human should validate every output or be held accountable in case of a mistake.
I take an interest in plane crashes and human factors in digital systems. We understand that there's a very human aspect of complacency that is often read about in reports of true disasters, well after that complacency has crept deep into an organization.
When you put something on autopilot, you also massively accelerate your process of becoming complacent about it -- which is normal, it is the process of building trust.
When that trust is earned but not deserved, problems develop. Often the system affected by complacency drifts. Nobody is looking closely enough to notice the problems until they become proto-disasters. When the human finally is put back in control, it may be to discover that the equilibrium of the system is approaching catastrophe too rapidly for humans to catch up on the situation and intercede appropriately. It is for this reason that many aircraft accidents occur in the seconds and minutes following an autopilot cutoff. Similarly, every Tesla that ever slammed into the back of an ambulance on the back of the road was a) driven by an AI, b) that the driver had learned to trust, and c) the driver - though theoretically responsible - had become complacent.
Theoretical? I don't see any reason that complacency is fine in science. If it's a high school science project and you don't actually care at all about the results, sure.
The problem is that the original statement is too black and white. We make tradeoffs based on costs and feasibility
"if the program underperforms compared to humans and starts making a large amount of errors, the human who set up the pipeline will be held accountable"
Like.. compared to one human? Or an army of a thousand humans tracking animals? There is no nuance at all. It's just unreasonable to make a blanket statement that humans always have to be accountable.
"If the program is responsible for a critical task .."
See how your statement has some nuance? and recognizes that some situations require more accountability and validation that others?
If some dogs chew up an important component, the CERN dog-catcher won't avoid responsibility just by saying "Well, the computer said there weren't any dogs inside the fence, so I believed it."
Instead, they should be taking proactive steps: testing and evaluating the AI, adding manual patrols, etc.
That doesn't follow. Say you write a proof for a something I request, I can then check that proof. That doesn't mean I don't derive any value from being given the proof. A lack of trust does not imply no use.
I'm also in Switzerland, currently my approach is to invest in Vanguard VOO (tracks the S&P500) via Interactive Brokers. There is a way to setup auto transfer and invest every month
As a caveat your money will be in dollars and in American companies, which might not be what you want, but it's worked for me well so far
I read this as "LLM-generated contributions" are not welcome, not "any contribution that used LLMs in any way".
More generally, this is clearly a rule to point to in order to end discussions with low effort net-negative contributors. I doubt it's going to be a problem for actually valuable contributions.
> Does this mean Copilot tab complete is banned too? What about asking an LLM for advice and then writing all the code yourself?
You're brushing up against some of the reasons why I am pretty sure policies like this will be futile. They may not diminish in popularity but they will be largely unenforceable. They may serve as an excuse for rejecting poor quality code or code that doesn't fit the existing conventions/patterns but did maintainers need a new reason to reject those PRs?
How does one show that no assistive technologies below some threshold were used?
That seems unlikely. Probably, what is going to happen, is if during a code review, you can't actually explain what your code is doing or why you wrote it, then you will be banned.
I don't know much about this project, but looking at the diff with their previous policy, it's pretty clear that people were abusing it and not declaring that they use llms, and they don't actually know what they're doing
Then a friendlier and clearer wording for the policy would work better. The current one says you will be "immediately" banned "without recourse" which implies that nothing like you are describing will happen.
Or arguably that's the point. If you Copilot generate a few lines of code or use it for inspiration you're still paying attention to it and are aware of what it's doing. The actual outcome will be indistinguishable from the code you hand wrote so it's fine. What policies like this do is stop someone generating whole pages at once, run it with minimal testing then chuck it into the code base forever.
I'm pretty sure the point is that anything clearly generated will result in an instant ban. That seems rather fair, you want contributors who only submit code they can fully understand and reason about.
The part you are quoting is being removed. The policy used to state "If you contribute un-reviewed LLM generated...", now simply states "If you use an LLM to make any kind of contribution then you will immediately be banned without recourse."
> Any contribution of any LLM-generated content will be rejected and result in an immediate ban for the contributor, without recourse.
You can argue it’s unenforceable, unproductive, or a bad idea. But it says nothing about unreviewed code. Any LLM generated code.
I’m not sure how great of an idea it is, but then again, it’s not my project.
Personally, I’d rather read a story about how this came to be. Either the owner of the project really hates LLMs or someone submitted something stupid. Either would be a good read.
Not sure about this project in particular, but many more popular projects (curl comes to mind) have adopted similar policies not out of spite but because they'd get submerged by slop.
Sure, a smart guy with a tool can do so much more, but an idiot with a tool can ruin it for everyone.
Isn't it then more reasonable to have a policy that "people who submit low quality PRs will be banned"? Target the actual problem rather than an unreliable proxy of the problem.
LLM-generated code can be high quality just as human-generated code can be low quality.
Also, having a "no recourse" policy is a bit hostile to your community. There will no doubt be people who get flagged as using LLMs when they didn't and denying them even a chance to defend themselves is harsh.
Banning LLMs can result in shorter arguments. "Low quality" is overly subjective and will probably take a lot of time to argue about. And then the possible outrage if it is taken to social media.
Can it really? "You submitted LLM-generated contributions" is also highly subjective. Arguably more so since you can't ever really be sure if somethingi s AI generated while with quality issues there are concrete things you can point to (e.g. it the code simply doesn't work, doesn't meet the contributor guidelines, uses obvious anti-patterns etc.).
If you rtfa, you will find it's actually the other way around. The linked PR from the AI has "concrete things you can point to" like "the code simply doesn't work".
I am wondering why you are posting this link, then asking this question to the HN community, instead of asking the project directly for more details.
I does look like your intent is to stir some turmoil over the project position, and not to contribute constructively to the project.
That kind of point could be made for a large fraction of HN comments, but that aside: if a project’s policy is to ban for any LLM usage, without recourse, just asking a question about it could put you on a list of future suspects…
Friendly reminder that em and en dashes were part of English well before ChatGPT was launched. Anecdotally, I’ve been using them forever and English isn’t even my native language.
Your example sounds like stopping notepad from rendering copyrighted content