5 ms·

The Record labels are far, far more litigious than the art community.

by LewisVerstappen 3y ago

PaulDavisThe1st 3y ago
They can't litigate a person doing this at home, and never redistributing. I suppose they might try, anyway.
- kmeisthax 3y ago
  The RIAA pioneered copyright enforcement at the individual level back in the 2000s, they absolutely would try to sue downstream AudioCraft users.
  - throwaway290 3y ago
    They should start with AudioCraft itself, conceptually it's derivative work and it doesn't matter if it's "open source" or not. Try throwing in someone's sample in a song and publish it saying "no copyright infringement intended and I totally don't make any money from it"... If it becomes popular, see how long it stays up until DMCA takedown. And we know this dataset is already popular.
    - PaulDavisThe1st 3y ago
      > and publish it This is precisely the opposite of the context I was remarking on.
      - throwaway290 3y ago
        AudioCraft itself is published. That's the context I am remarking on.
  - PaulDavisThe1st 3y ago
    [flagged]
    - manquer 3y ago
      Even before streaming, You never "owned" any music legally [1], you merely owned a physical copy of a performance[3] of a song, that in no way gives you the right to make derivative works [2] automatically. Also it doesn't really matter on what the law says, RIAA in the last iteration, relied on the fact it you would rather pay a fine than be able to afford expensive lawyers to fight the specifics out in court on average. It was always about disproportionate ability to bring resources against individual "offenders" to create fear among everyone to deter "undesirable" forms of copying, not necessarily what the legal protections were. --- [1] Unless you specifically commissioned it under a contract which gave you the right [2] See recent cases including those related to Kris Kashtanova and Andy Warhol. [3] Not the song, just the performance, aka Taylor Swift version, for good explanation of how the rights are divvied up in the music industry a Planet Money series covers it well https://www.npr.org/sections/money/2022/10/29/1131927591/inflation-song-music-label-earnest-jackson https://www.npr.org/sections/money/2022/10/29/1131927591/inf...
      - PaulDavisThe1st 3y ago
        Establishing derivation is at the crux of all legal matters surrounding diffusion models. It has not yet been clearly established. If it is, then I'd agree with you. Until then, I think it's a bit more up in the air. Also, IIRC, RIAA did not bring many resources to bear against e.g. "home taping" itself, because they could essentially never know that it had occured. The overwhelming majority of their efforts went into trying to takedown people distributing multiple copies. The Kashtanova case does not cover derivation in any real way, but is really about copyright attribution choices between human and software. The Warhol case specifically tests a fair use claim, not a derivation claim.
        cmgbhm 3y ago
        https://en.wikipedia.org/wiki/Audio_Home_Recording_Act https://en.wikipedia.org/wiki/Audio_Home_Recording_Act https://en.wikipedia.org/wiki/Home_Taping_Is_Killing_Music https://en.wikipedia.org/wiki/Home_Taping_Is_Killing_Music Recording industries have fought end user reproduction often. They’ve fought sampling battles. Go after the pocketbooks and go after the technology waves. If there’s a derivative argument they can make, they will.
- jsheard 3y ago
  Is training a "pirate model" something you'd reasonably be able to do at home though, given the compute requirements? The analogous "image generation at home" is only possible due to a for-profit entity with significant resources choosing to (a) play fast-and-loose with the provenance of their training set and (b) giving away the resulting model for free, if the open source community had to train their models from scratch then as best as I can tell they would still be stuck in the dark ages generating vague goopy abominations.
  - throwuwu 3y ago
    You could take a model trained on CC content and then fine tune it on copyrighted material cheaply and quickly
  - PaulDavisThe1st 3y ago
    Currently, yes, available compute power @ home does indeed seem like a limitation. Whether that remains true going forward seems a little unclear to me.

The Record labels are far, far more litigious than the art community.

PaulDavisThe1st 3y ago

They can't litigate a person doing this at home, and never redistributing. I suppose they might try, anyway.

kmeisthax 3y ago
The RIAA pioneered copyright enforcement at the individual level back in the 2000s, they absolutely would try to sue downstream AudioCraft users.
- throwaway290 3y ago
  They should start with AudioCraft itself, conceptually it's derivative work and it doesn't matter if it's "open source" or not. Try throwing in someone's sample in a song and publish it saying "no copyright infringement intended and I totally don't make any money from it"... If it becomes popular, see how long it stays up until DMCA takedown. And we know this dataset is already popular.
  - PaulDavisThe1st 3y ago
    > and publish it This is precisely the opposite of the context I was remarking on.
    - throwaway290 3y ago
      AudioCraft itself is published. That's the context I am remarking on.
- PaulDavisThe1st 3y ago
  [flagged]
  - manquer 3y ago
    Even before streaming, You never "owned" any music legally [1], you merely owned a physical copy of a performance[3] of a song, that in no way gives you the right to make derivative works [2] automatically. Also it doesn't really matter on what the law says, RIAA in the last iteration, relied on the fact it you would rather pay a fine than be able to afford expensive lawyers to fight the specifics out in court on average. It was always about disproportionate ability to bring resources against individual "offenders" to create fear among everyone to deter "undesirable" forms of copying, not necessarily what the legal protections were. --- [1] Unless you specifically commissioned it under a contract which gave you the right [2] See recent cases including those related to Kris Kashtanova and Andy Warhol. [3] Not the song, just the performance, aka Taylor Swift version, for good explanation of how the rights are divvied up in the music industry a Planet Money series covers it well https://www.npr.org/sections/money/2022/10/29/1131927591/inflation-song-music-label-earnest-jackson https://www.npr.org/sections/money/2022/10/29/1131927591/inf...
    - PaulDavisThe1st 3y ago
      Establishing derivation is at the crux of all legal matters surrounding diffusion models. It has not yet been clearly established. If it is, then I'd agree with you. Until then, I think it's a bit more up in the air. Also, IIRC, RIAA did not bring many resources to bear against e.g. "home taping" itself, because they could essentially never know that it had occured. The overwhelming majority of their efforts went into trying to takedown people distributing multiple copies. The Kashtanova case does not cover derivation in any real way, but is really about copyright attribution choices between human and software. The Warhol case specifically tests a fair use claim, not a derivation claim.
      - cmgbhm 3y ago
        https://en.wikipedia.org/wiki/Audio_Home_Recording_Act https://en.wikipedia.org/wiki/Audio_Home_Recording_Act https://en.wikipedia.org/wiki/Home_Taping_Is_Killing_Music https://en.wikipedia.org/wiki/Home_Taping_Is_Killing_Music Recording industries have fought end user reproduction often. They’ve fought sampling battles. Go after the pocketbooks and go after the technology waves. If there’s a derivative argument they can make, they will.
jsheard 3y ago
Is training a "pirate model" something you'd reasonably be able to do at home though, given the compute requirements? The analogous "image generation at home" is only possible due to a for-profit entity with significant resources choosing to (a) play fast-and-loose with the provenance of their training set and (b) giving away the resulting model for free, if the open source community had to train their models from scratch then as best as I can tell they would still be stuck in the dark ages generating vague goopy abominations.
- throwuwu 3y ago
  You could take a model trained on CC content and then fine tune it on copyrighted material cheaply and quickly
- PaulDavisThe1st 3y ago
  Currently, yes, available compute power @ home does indeed seem like a limitation. Whether that remains true going forward seems a little unclear to me.