6 ms·

> "It's clear some people have had their brain broken by the existence of AI." The AI wrote code which worked, for a problem the submitter had, which had not b

> "It's clear some people have had their brain broken by the existence of AI." The AI wrote code which worked, for a problem the submitter had, which had not been solved by any human for a long time, and there is limited human developer time/interest/funding available for solving it. Dumping a mass of code (and work) onto maintainers without prior discussion is the problem[1]. If they had forked the repo, patched it themselves with that PR and used it personally, would they have a broken brain because of AI? They claim to have read the code, tested the code, they know that other people want the functionality; is wanting to share working code a "broken brain"? If the AI code didn't work - if it was slop - and they wanted the maintainers to fix it, or walk them through every step of asking the AI to fix it - that would be a huge problem, but that didn't happen. [1] copyrightwashing and attribution is another problem but also not one that's "broken brain by the existence of AI" related.

nsagent 11mo ago
It's clear Claude adapted code directly from the OxCaml implementation (the PR author said he pointed Claude at that code [1] and then provides a ChatGPT analysis [2] that really highlights the plagiarism, but ultimately comes to the conclusion that it isn't plagiarized). Either that highlights someone who is incompetent or they are willfully being blasé. Neither bodes well for contributing code while respecting copyright (though mixing and matching code on your own private repo that isn't distributed in source or binary form seems reasonable to me). [1]: https://github.com/ocaml/ocaml/pull/14369#issuecomment-3557357958 https://github.com/ocaml/ocaml/pull/14369#issuecomment-35573... [2]: https://github.com/ocaml/ocaml/pull/14369#issuecomment-3556624486 https://github.com/ocaml/ocaml/pull/14369#issuecomment-35566...
- menaerus 11mo ago
  Sorry, this is just ridicilous and shows how people fragile really are. This whole topic and whole MR as well. I am routinely looking into the folly implementation, sometimes into the libstdc++, sometimes into libc++, sometimes into boost or abseil etc. to find inspiration for problems that I tackle in other codebases. By the same standards, this should also be plagiarism, no? I manufacture new ideas by compiling existing knowledge from elsewhere. Literally every engineer in the world does the same. Why is AI any different?
  - meheleventyone 11mo ago
    Perhaps because the AI assigned copyright in the files to the author of the library it copied from and the person prompting it told it to look at that library. Without even getting into the comedy AI generated apologia to go with it which makes it look worse rather than better. From a pragmatic viewpoint as an engineer you assign the IP you create over to the company you work for so plagarism has real world potential to lose you your job at best. There's a difference between taking inspiration from something unrelated "oh this is a neat algorithmic approach to solving this class of problems" to "I need to implement this specific feature and it exists in this library so I'll lift it nearly verbatim".
    - menaerus 11mo ago
      Can you give an example what exactly was copied? I ask because I took a look into MR and original repo, and the conclusion is that the tool only copy-pasted the copyright header but not the code. So I am still wondering - what's wrong with that (it's a silly mistake even a human can make), and where is the copyright infringement everyone is talking about?
      - DrammBA 11mo ago
        > copy-past[ing] the copyright header but not the code [is] a silly mistake even a human can make Do you mind showing me some examples of that? That seems so implausible to me Just for reference, here's another example of AI adding phantom contributors and the human just ignoring it or not even noticing: https://github.com/auth0/nextjs-auth0/issues/2432 https://github.com/auth0/nextjs-auth0/issues/2432
        menaerus 11mo ago
        > Do you mind showing me some examples of that? That seems so implausible to me What's so special about it that I need to show you the example?
        11mo ago
- joelreymont 11mo ago
  The key is that AI adapted, not stole. It's actually capable of reasoning and generating derivative code and not just copying stuff wholesale. See examples at the bottom of my post: https://joel.id/ai-will-write-your-next-compiler/ https://joel.id/ai-will-write-your-next-compiler/
  - noticingdecline 11mo ago
    [dead]
torben-friis 11mo ago
>They claim to have read the code, tested the code, they know that other people want the functionality; is wanting to share working code a "broken brain"? There is clearly a deviation between the amount of oversight the author thinks they provided and the actual level of oversight. This is clear by the fact that they couldn’t even explain the misattribution. They also mention that this is not their area of expertise. In general, I think that it is a reasonable assumption that, if you couldn’t have written the code yourself, you’re in no position to claim you can ensure its quality.
- jodrellblank 11mo ago
  If a manager says they provided oversight of their developer employees, and the code was not as good as the manager thought, would you say "the manager has had their brain broken by the existence of employees"?
  - palmotea 11mo ago
    > If a manager says they provided oversight of their developer employees, and the code was not as good as the manager thought, would you say "the manager has had their brain broken by the existence of employees"? That could be either regular incompetence or a "broken brain." It's more towards latter if the manager had no clue about what was going on, even after having it explained to him. This guy is equivalent to a manager who hired two bozos to do a job, but insisted it was good because he had them check each other's work and what they made didn't immediately fall down.
  - torben-friis 11mo ago
    I'll bite, let's grant for the sake of the argument that equaling the LLM with a person holds. This manager is directly providing an unrelated team with an unprompted 150-file giant PR dumped at once with no previous discussion. Upon questioning, he says the code has been written by an outside contractor he personally chose. No one has onboarded this contractor to the team, and checking their online presence shows lots of media appearances, but not a single production project in their CV, much less long time maintenance. A cursory glance at the files reveals that the code contains copypasted code from stackoverflow to the point that the original author's name is still pasted in comments. The manager can not justify this, but doesn't seem bothered by the fact, and insists that the contractors are amazing because he's been following them in social networks and is infatuated with their media there. Furthermore, you check the manager's history in slack and you see 15 threads of him doing the same for other teams. The ones that have agreed to review their PRs have closed them for being senseless. How would you be interacting with this guy?
ath3nd 11mo ago
[dead]
torben-friis 11mo ago
Also, I just took a glance at the PR and even without knowing the language it took 10 seconds for the first WTF. The .md documents Claude generated are added to .gitignore, including one for the pr post itself. That’s the quality he’s vouching for.
arresin 11mo ago
You are under a delusion. I’m serious.
joelreymont 11mo ago
People complaining about AI stealing code may not realize that OxCaml is a fork of the code that AI is modifying. Both forks have the same license and there are people working on both projects. AI did paste Mark's copyright on all the files for whatever reason but it did not lift the DWARF implementation from OxCaml and paste it into the PR. The PR wasn't one-shot either. I had to steer it to completion over several days, had one AI review the changes made by the other, etc. The end result is that the code _works_ and does what it says on the tin!

nsagent 11mo ago

It's clear Claude adapted code directly from the OxCaml implementation (the PR author said he pointed Claude at that code [1] and then provides a ChatGPT analysis [2] that really highlights the plagiarism, but ultimately comes to the conclusion that it isn't plagiarized). Either that highlights someone who is incompetent or they are willfully being blasé. Neither bodes well for contributing code while respecting copyright (though mixing and matching code on your own private repo that isn't distributed in source or binary form seems reasonable to me). [1]: https://github.com/ocaml/ocaml/pull/14369#issuecomment-3557357958 https://github.com/ocaml/ocaml/pull/14369#issuecomment-35573... [2]: https://github.com/ocaml/ocaml/pull/14369#issuecomment-3556624486 https://github.com/ocaml/ocaml/pull/14369#issuecomment-35566...

menaerus 11mo ago
Sorry, this is just ridicilous and shows how people fragile really are. This whole topic and whole MR as well. I am routinely looking into the folly implementation, sometimes into the libstdc++, sometimes into libc++, sometimes into boost or abseil etc. to find inspiration for problems that I tackle in other codebases. By the same standards, this should also be plagiarism, no? I manufacture new ideas by compiling existing knowledge from elsewhere. Literally every engineer in the world does the same. Why is AI any different?
- meheleventyone 11mo ago
  Perhaps because the AI assigned copyright in the files to the author of the library it copied from and the person prompting it told it to look at that library. Without even getting into the comedy AI generated apologia to go with it which makes it look worse rather than better. From a pragmatic viewpoint as an engineer you assign the IP you create over to the company you work for so plagarism has real world potential to lose you your job at best. There's a difference between taking inspiration from something unrelated "oh this is a neat algorithmic approach to solving this class of problems" to "I need to implement this specific feature and it exists in this library so I'll lift it nearly verbatim".
  - menaerus 11mo ago
    Can you give an example what exactly was copied? I ask because I took a look into MR and original repo, and the conclusion is that the tool only copy-pasted the copyright header but not the code. So I am still wondering - what's wrong with that (it's a silly mistake even a human can make), and where is the copyright infringement everyone is talking about?
    - DrammBA 11mo ago
      > copy-past[ing] the copyright header but not the code [is] a silly mistake even a human can make Do you mind showing me some examples of that? That seems so implausible to me Just for reference, here's another example of AI adding phantom contributors and the human just ignoring it or not even noticing: https://github.com/auth0/nextjs-auth0/issues/2432 https://github.com/auth0/nextjs-auth0/issues/2432
      - menaerus 11mo ago
        > Do you mind showing me some examples of that? That seems so implausible to me What's so special about it that I need to show you the example?
        11mo ago
joelreymont 11mo ago
The key is that AI adapted, not stole. It's actually capable of reasoning and generating derivative code and not just copying stuff wholesale. See examples at the bottom of my post: https://joel.id/ai-will-write-your-next-compiler/ https://joel.id/ai-will-write-your-next-compiler/
- noticingdecline 11mo ago
  [dead]

torben-friis 11mo ago

>They claim to have read the code, tested the code, they know that other people want the functionality; is wanting to share working code a "broken brain"? There is clearly a deviation between the amount of oversight the author thinks they provided and the actual level of oversight. This is clear by the fact that they couldn’t even explain the misattribution. They also mention that this is not their area of expertise. In general, I think that it is a reasonable assumption that, if you couldn’t have written the code yourself, you’re in no position to claim you can ensure its quality.

jodrellblank 11mo ago
If a manager says they provided oversight of their developer employees, and the code was not as good as the manager thought, would you say "the manager has had their brain broken by the existence of employees"?
- palmotea 11mo ago
  > If a manager says they provided oversight of their developer employees, and the code was not as good as the manager thought, would you say "the manager has had their brain broken by the existence of employees"? That could be either regular incompetence or a "broken brain." It's more towards latter if the manager had no clue about what was going on, even after having it explained to him. This guy is equivalent to a manager who hired two bozos to do a job, but insisted it was good because he had them check each other's work and what they made didn't immediately fall down.
- torben-friis 11mo ago
  I'll bite, let's grant for the sake of the argument that equaling the LLM with a person holds. This manager is directly providing an unrelated team with an unprompted 150-file giant PR dumped at once with no previous discussion. Upon questioning, he says the code has been written by an outside contractor he personally chose. No one has onboarded this contractor to the team, and checking their online presence shows lots of media appearances, but not a single production project in their CV, much less long time maintenance. A cursory glance at the files reveals that the code contains copypasted code from stackoverflow to the point that the original author's name is still pasted in comments. The manager can not justify this, but doesn't seem bothered by the fact, and insists that the contractors are amazing because he's been following them in social networks and is infatuated with their media there. Furthermore, you check the manager's history in slack and you see 15 threads of him doing the same for other teams. The ones that have agreed to review their PRs have closed them for being senseless. How would you be interacting with this guy?

ath3nd 11mo ago

[dead]

torben-friis 11mo ago

Also, I just took a glance at the PR and even without knowing the language it took 10 seconds for the first WTF. The .md documents Claude generated are added to .gitignore, including one for the pr post itself. That’s the quality he’s vouching for.

arresin 11mo ago

You are under a delusion. I’m serious.

joelreymont 11mo ago

People complaining about AI stealing code may not realize that OxCaml is a fork of the code that AI is modifying. Both forks have the same license and there are people working on both projects. AI did paste Mark's copyright on all the files for whatever reason but it did not lift the DWARF implementation from OxCaml and paste it into the PR. The PR wasn't one-shot either. I had to steer it to completion over several days, had one AI review the changes made by the other, etc. The end result is that the code _works_ and does what it says on the tin!