4 ms·

It's true no project wants that type of "contributions." But this policy also bans long-time contributors from thoughtful use of LLM-generated code.

by loeg 3mo ago

attila-lendvai 3mo ago
i believe the reasoning behind this is of legal nature (copyright/copyleft, etc).
rswail 3mo ago
LLM-generated code is not subject to copyright.
- lowbloodsugar 3mo ago
  Unless there is significant human input and guidance. But IANAL.
- jdub 3mo ago
  That is not an established precedent. (It's not what that case decided.)
  - dspillett 3mo ago
    But neither is the opposite.
    - jdub 3mo ago
      "that is not an established precedent" in response to a specific claim and "there is no established precedent" can both be true without laying down a verbose foundation of caveats
  - 20k 3mo ago
    Its not specifically established but its the only compatible ruling that the law can really make here. There is a lot of very similar precedence, there was a legaleagle on this which went through the likely legal outcome
    - jdub 3mo ago
      Given a nation* that decides "corporations are people" and "money is speech", let's not be unduly optimistic about (even temporary) legislation or interpretation. * focusing on the US given this site and the outsized influence it has on global copyright
      - rswail 3mo ago
        Corporate personhood is a common legal practice, where necessary the law distinguishes between "persons" and "natural persons". That's the same in the US/UK/AU/NZ/CA/DE/JP/IN. The difference in the US is that SCOTUS basically made it impossible to restrict corporate spending on politics. That is what needs to be fixed in the US, not Corporate Personhood. https://en.wikipedia.org/wiki/Corporate_personhood https://en.wikipedia.org/wiki/Corporate_personhood
josefx 3mo ago
It explicitly allows research, tests and third party tools. I am sure that any contributer can still find plenty of use for AI with that.
- moomin 3mo ago
  Indeed, one useful thing you can do is fire questions like "I've got three different solutions to this problem, which performs best?". You don't need to keep the code, you just want to know which direction you should take.
flufluflufluffy 3mo ago
I’d be very surprised if any long time contributor to gcc would want to start pushing AI generated code. But even if that is the case, this is one of the largest and most complex pieces of software on the planet, and is used to produce the binaries of countless other pieces of software. I think this policy makes sense because of that. We want to make sure no inadvertent behavior is introduced (whatever that might be, from bugs to ~puts conspiracy theory hat on~ some kind of backdoor undetectable by humans inserted into every compiled binary allowing the authoring AI system (or the government behind it) remote access to every machine it runs on). Okay maybe that’s a little crazy
- graemep 3mo ago
  > But even if that is the case, this is one of the largest and most complex pieces of software on the planet, and is used to produce the binaries of countless other pieces of software. I think this policy makes sense because of that. We want to make sure no inadvertent behavior is introduced That would be an excellent reason to limit AI contributions if true. I am not in a position to judge the requirements of a compiler project so I do not have an opinion on whether that is a sound reason or not. However, the reasons being given for the change (and the wording around "legally significant") seem to be about copyright law which I do not think is sound reasoning (see my other comments on this). > ~puts conspiracy theory hat on~ some kind of backdoor undetectable by humans inserted into every compiled binary allowing the authoring AI system (or the government behind it) remote access to every machine it runs on). Okay maybe that’s a little crazy Someone using AI code to introduce a back door will surely be willing to lie and say its human written so this policy will not help.
- atq2119 3mo ago
  As a non-GCC compiler developer myself, there's a spectrum of AI use and the words "AI generated code" can mean different things to different people. At the extreme ends, nobody I know well objects to AI auto-completion in your editor, but neither does anybody want to let AI open pull requests autonomously. Having a coding agent iterate on a well-defined narrow task in the background while I'm in a meeting or working on something else, and then doing a thorough, thoughtful, and active local code review (meaning, having an editor and the diff open side-by-side and liberally making edits to clean up the code) before pushing anything out for others to see feels like a reasonable point in the space right now. Nothing is perfect, but this style of active code review removes basically all AI smells in practice. The edit-build-test loop is fairly long (i.e., longer than a few seconds), so having AI babysit this loop for the initial development of a task is helpful.

It's true no project wants that type of "contributions." But this policy also bans long-time contributors from thoughtful use of LLM-generated code.

by loeg 3mo ago

It's true no project wants that type of "contributions." But this policy also bans long-time contributors from thoughtful use of LLM-generated code.

attila-lendvai 3mo ago

i believe the reasoning behind this is of legal nature (copyright/copyleft, etc).

rswail 3mo ago

LLM-generated code is not subject to copyright.

lowbloodsugar 3mo ago
Unless there is significant human input and guidance. But IANAL.
jdub 3mo ago
That is not an established precedent. (It's not what that case decided.)
- dspillett 3mo ago
  But neither is the opposite.
  - jdub 3mo ago
    "that is not an established precedent" in response to a specific claim and "there is no established precedent" can both be true without laying down a verbose foundation of caveats
- 20k 3mo ago
  Its not specifically established but its the only compatible ruling that the law can really make here. There is a lot of very similar precedence, there was a legaleagle on this which went through the likely legal outcome
  - jdub 3mo ago
    Given a nation* that decides "corporations are people" and "money is speech", let's not be unduly optimistic about (even temporary) legislation or interpretation. * focusing on the US given this site and the outsized influence it has on global copyright
    - rswail 3mo ago
      Corporate personhood is a common legal practice, where necessary the law distinguishes between "persons" and "natural persons". That's the same in the US/UK/AU/NZ/CA/DE/JP/IN. The difference in the US is that SCOTUS basically made it impossible to restrict corporate spending on politics. That is what needs to be fixed in the US, not Corporate Personhood. https://en.wikipedia.org/wiki/Corporate_personhood https://en.wikipedia.org/wiki/Corporate_personhood

josefx 3mo ago

It explicitly allows research, tests and third party tools. I am sure that any contributer can still find plenty of use for AI with that.

moomin 3mo ago
Indeed, one useful thing you can do is fire questions like "I've got three different solutions to this problem, which performs best?". You don't need to keep the code, you just want to know which direction you should take.

flufluflufluffy 3mo ago

I’d be very surprised if any long time contributor to gcc would want to start pushing AI generated code. But even if that is the case, this is one of the largest and most complex pieces of software on the planet, and is used to produce the binaries of countless other pieces of software. I think this policy makes sense because of that. We want to make sure no inadvertent behavior is introduced (whatever that might be, from bugs to ~puts conspiracy theory hat on~ some kind of backdoor undetectable by humans inserted into every compiled binary allowing the authoring AI system (or the government behind it) remote access to every machine it runs on). Okay maybe that’s a little crazy

graemep 3mo ago
> But even if that is the case, this is one of the largest and most complex pieces of software on the planet, and is used to produce the binaries of countless other pieces of software. I think this policy makes sense because of that. We want to make sure no inadvertent behavior is introduced That would be an excellent reason to limit AI contributions if true. I am not in a position to judge the requirements of a compiler project so I do not have an opinion on whether that is a sound reason or not. However, the reasons being given for the change (and the wording around "legally significant") seem to be about copyright law which I do not think is sound reasoning (see my other comments on this). > ~puts conspiracy theory hat on~ some kind of backdoor undetectable by humans inserted into every compiled binary allowing the authoring AI system (or the government behind it) remote access to every machine it runs on). Okay maybe that’s a little crazy Someone using AI code to introduce a back door will surely be willing to lie and say its human written so this policy will not help.
atq2119 3mo ago
As a non-GCC compiler developer myself, there's a spectrum of AI use and the words "AI generated code" can mean different things to different people. At the extreme ends, nobody I know well objects to AI auto-completion in your editor, but neither does anybody want to let AI open pull requests autonomously. Having a coding agent iterate on a well-defined narrow task in the background while I'm in a meeting or working on something else, and then doing a thorough, thoughtful, and active local code review (meaning, having an editor and the diff open side-by-side and liberally making edits to clean up the code) before pushing anything out for others to see feels like a reasonable point in the space right now. Nothing is perfect, but this style of active code review removes basically all AI smells in practice. The edit-build-test loop is fairly long (i.e., longer than a few seconds), so having AI babysit this loop for the initial development of a task is helpful.