8 ms·
EFF to Courts: Don't Rewrite Copyright over AI Hype
- josefritzishere 1mo agoThis is solid guidance. "Given this range of experimentation, courts should be reluctant to decide in advance what tools do and do not foster “human creativity.”"
- boxed 1mo agoThere might need to be some calibration. The reason for copyright is (ostensibly) to compensate for the enormous cost of production. If the cost goes down, the protection should also logically go down.
- LtWorf 1mo agoThe cost of producing decent things hasn't gone down. Also the cost of slop is not that low, just subsidised.
- boxed 1mo agoI'm a programmer by trade. I know you're wrong from much personal experience, and the difference between last year and today is dramatic. If you believe the anti-AI people you will become misinformed, just as much as if you listen to the most pro-AI people.
- LtWorf 1mo ago> I'm a programmer by trade. Me too. I don't understand what makes you think your opinion is superior to all the evidence I've seen so far.
- boxed 1mo agoBecause the evidence only goes in one direction. If I can get substantially more work done because of AI tools that falsifies the idea that LLMs don't help. Same as a satellite orbiting the earth once with a camera filming falsifies flat earthism. It doesn't matter how much evidence there is to say the earth is flat, it's falsified.
- LtWorf 1mo agoExcept all scientific research points in the other direction so, my dear professional developer, in this situation you're the flat earther :)
- HNDevsSuck 1mo ago[dead]
- repelsteeltje 1mo agoI don't think the reason for copyright law is compensation, per se. It's ownership: being in a position where the creator is able to decide what to do with it. Compensation being one option, but also restricting use our controlling access.
- bediger4000 1mo agoThat's the US constitutional reason. Give a creator a temporary monopoly so as to compensate them for ultimately making the public domain larger.
- boxed 1mo agoOwnership is the proxy they could figure out. It's not more complicated than that.
- msabalau 1mo agoThe purpose of copyright under the US Constitution, a text relevant to the courts the EFF is addressing here, is to "promote the progress of science and useful arts by giving creators temporary exclusive rights to their works" It is not to "compensate for the cost of production", nor is it to ensure that business plans of major music publishers happen to be successful. The EFF is correct to say here that copyright shouldn't be perverted to limiting market competition generally, but limited to protecting a specific works.
- bpodgursky 1mo agoWhy are you being argumentative? "Promote" and "exclusive rights" was always about compensation for the time and effort invested. You just rephrased what the OP said while claiming to disagree.
- bux93 1mo agoIt's substantively different. The level of compensation you could expect from copyright needs to be high enough that creators/inventors in general feel that it may be worthwhile to create/invent stuff. That level of compensation may be exactly enough to pay for a musician to make a track. We know this is not the case, it's more like a lottery ticket; some musicians make millions (much more than would be needed to entice them to choose this line of work) and many make almost nothing, but still persist. If you relax copyright provisions so that average compensation is halved, this would presumably have a non-linear impact. Maybe a 50% decrease in expected compensation only results in -10% new works. Maybe in -80%. For example, it is completely implausible that retroactively extending copyright (as has been done) results in any additional works being created at all - just higher profits for existing owners. This also applies to the costs; a higher cost of production (e.g. cost of living) doesn't necessarily need to lead to more stringent copyright restraints, nor lower costs to more lax restrictions.
- boxed 1mo agoI don't get how what you said is different from what I said.
- 1mo ago
- bogwog 1mo agoCopyright has nothing to do with cost of production. What does it cost for me to doodle something on a napkin? Nothing, yet I still enjoy the full protections copyright law offers for that doodle (with the exception today being AI companies. Apparently, they can do whatever the fuck they want with it)
- cestith 1mo agoDo you mean production or reproduction? For instance, have you ever looked and the budget or the credits for a major film, TV, or game project?
- anonyfox 1mo agoIndependend of AI I am fully supporting the idea to abolish copyright in general, full stop. Allowing cheap duplication and tweaking is what leads to fast innovation and culture changes. Explicit (costly) patents can remain in place to secure industrial-grade research processes, but the general everyday stuff, incluing music and literature and code and so on is something to strip default rights away to allow faster progress. and oh the poor artists: next to none percentage wise gets much royalties anymore, streaming income is close to zero usually even if popular, the only real world industry to defend are the parasites that do not create art to begin with but siphon production/distribution channels. you might disagree with me, just wanna make clear that some people like me are against this copyright theater before AI blew up in popularity.
- andsoitis 1mo ago> fast innovation and culture changes Does this mean you think innovation and change in our culture is too slow? What’s you’re yardstick?
- inglor_cz 1mo agoHmm. Looking at the cinemas, almost all the top movies are prequels or sequels, milking the same topic for the tenth time. I'd say that at least the movie industry is somehow stuck in a rut.
- paulddraper 1mo agoGetting rid of copyright is going to make that problem worse not better... I'm wondering if you are reasoning correctly about this.
- inglor_cz 1mo agoI'd say that random people, if they reuse a copyrighted material, are a lot more likely to veer far off the beaten path than the big studios. Fan fiction, even if building on previous materials, seems to be much less repetitive to me.
- ThePhysicist 1mo agoTo be fair AI as we have it today wouldn't exist if everything was copyrighted and copyright was enforced very strictly across the web. Not sure if it's fair use and I get that it's not great if these companies try to monopolize knowledge (if they even can do that) but overall being more lax on copyright for fair use is good. Especially scientific knowledge should be openly accessibly to anyone, imagine what we could achieve if we could train LLMs on everything and make it accessible to everyone.
- gwbas1c 1mo agoI think we need to adjust copyright to balance market needs and preservation needs. AI works best when we can strike a balance between the needs of the creator to get compensated and the need of the general public to use content in ways the creator didn't anticipate.
- bluGill 1mo agoYou might be right, but I think we need at least 10 years before we can really start this discussion, and the discussion itself needs to take many more years.
- paimapi 1mo agoin the few books I've read about technological breakthroughs, the vast majority come from academia. I'm thinking about Eckert's adoption of IBM's ticker tape for commands rather than just rote data [0], all the inventions that came out of the NDRC [1] (for better or worse), the ENIAC, FORTRAN, how reliant solid state devices were on understanding quantum theory [2], and so many other examples. Paul E. Ceruzzi's "Computing: A Concise History" goes over much of this (including making note of the US military's heavy usage of LLMs for the Iraq War pre-empted the consumer push for LLMs) most of the things that private enterprise contributes is cost reduction and manufacturing capacity (important in their own right for ubiquity) but having extremely well-funded R&D and universities still seem like the primary driver for truly evolutionary and revolutionary breakthroughs rather than useful but not necessary optimizations IP law very much aims to protect the latter so that breakthroughs can be capitalized on but aren't a necessary motivator for researchers to continue frontier work (0)https://en.wikipedia.org/wiki/IBM_SSEC https://en.wikipedia.org/wiki/IBM_SSEC (1)https://en.wikipedia.org/wiki/National_Defense_Research_Committee https://en.wikipedia.org/wiki/National_Defense_Research_Comm... (2)https://www.allaboutcircuits.com/textbook/semiconductors/chpt-2/quantum-devices/ https://www.allaboutcircuits.com/textbook/semiconductors/chp...
- AlexandrB 1mo agoCopyright in general, is too long, too broad, and - thanks largely to DMCA - enforced too strictly. One reason for China's success in tech is their willingness to ignore all the IP nonsense (or at least look the other way) allowing folks to rapidly iterate on ideas without running into IP rent-seekers at every turn.
- Roark66 1mo agoI recently heard a local EU politician on the radio trying to convince the listeners "things made with AI should be inherently non copyrightable, because they lack human creative labour". I wish I could tell him there is this thing called photography where all a human does is presses a button and copyrightable artefacts are created by capturing how the world looks. For example a landscape that human had no influence over creating! Surely this "non copyrightability" should apply to photography too. /sarcasm AI is just a tool. You can make slop or you can make beautiful art. Is it fair to Picasso's copyright holder that all his paintings were used for training and you can prompt "this in style of Picasso"? No, especially if the company that trained the LLM makes money on it. But there is a technical solution. Simply during training record who owns what training data and record how much influence on each weight this training run had. Then during inference it could be recorded if the weights that trained on Picasso's stuff activated or not. And if someone pays for inference the original owner of tgat data should get a cut. Maybe even blockchain tech could be used for this... Imagine the hype! AI and crypto together!
- Ekaros 1mo agoI consider current length of copyright to be extremely too long. On other hand I also think that creators should be compensated when their creations are used to train AI models and this compensation should be substantial per company. And to any fallacious person telling about humans also "learning". Well humans brains are not replicated en masse just yet. And single learned person is not made to millions of copies and made to answer inane queries day and night...
- beej71 1mo agoI disagree with restrictions on digestive use of a work. If I legally acquire a book, I should be able to run that data through an algorithm, and I should be allowed to share the result freely as long as it's not replicating the original.
- Ekaros 1mo agoMy biggest contention is the acquiring price for such use. I hold that for such use the cost such be significantly higher than just if you were buying book for reading.
- beej71 1mo agoI see setting such pricing to be an impossible task. What should be the fee for me sharing the computed word count of a book, an uncopyrightable figure? We'd likely agree that a large payment should not be required in that case. And now we have to somehow legally define all these shades of gray, statistics, book summaries—a mess.
- 123aGa 1mo agoThe EFF is in the pocket of Big-AI, and this isn't the first time. "It is not the place of courts to say these people are not artists or that AI cannot augment human creativity in a positive way." Yes it is. If you had ever created anything, which you have not, you would know! You are a couple of tech bureaucrats who want foundation salaries and change with the political winds.
- ChannelFence 1mo ago[dead]
- tancop 1mo agoWe need to limit copyright to commercial use. Piracy only hurts corporate profits (CDPR proved it will never bankrupt a studio) but a media company getting an artist to work for them and not paying is seriously unfair. This kind of breaks down with independent artists but any economic loss for them would be way less important than the benefits of a system where all culture is shared and open for modification. We can't put the interests of a small group over everyone else if the members of that group choose to be in it.
- preisschild 1mo agoAnd what if that actually leads to less open-ness? If I license code under a copyleft license, users are forced to give back their changes to the public. Without copyright they would not be forced to do this, which means less work will be published.
- bogwog 1mo agoGuess it's time to stop donating to the EFF. They're completely off the mark this time. The random loser generating slop in his bedroom isn't necessarily the one infringing, it's the AI company that took all of our work and is reselling it that is infringing. This is an obvious blind spot in copyright law and no reasonable person would consider that to be fair use. The challenge I think is communicating to a layman how AI training works. For the people currently in the grips of AI psychosis: new regulation here is good for AI too. AI models will not get more creative without new human training data, but the current exploitative situation means people are not incentivized to keep creating. Human artists who create new things should be rewarded by the market, as copyright law intended. Even if their work only ever ends up feeding some gen AI model, they should still be entitled to compensation in some way (eg royalties). They worked hard so that you don't have to.™ Instead, AI companies are the ones being rewarded for it. That's not sustainable for anyone.
- seki285 1mo agoIt's funny to hear people repeat, "AI will replace you!" Then beg for more training data of humans being humans.
- fny 1mo agoUnder the current state of affairs, copyleft material can be laundered through LLMs without triggering proprietary weights from becoming open. Something feels broken.
- bluGill 1mo agoYou could always do that with copyright programs. The only thing that changed was the cost. The only example of somebody doing this was with the BIOS of the original IBM PC, but it was always possible and since nothing was copied it is legal. Somebody else did all the hard work. Patents are the only protection that should exist for this type of thing. (though finding an algorithm worth patenting is hard - most are either a small variation easy to work around, or so old that prior art exists from 50 years ago).
- sandinmyjoints 1mo agoIs "only" doing a lot of work in your second sentence? Sometimes a change in degree is a change in kind.
- LtWorf 1mo agoYep. I've seen it happen and there's basically no recourse against it currently. And EFF is saying that it's fine… honestly not what I was expecting to read.
- Lucaichyexp 1mo ago[flagged]
- alphadelphi 1mo agolegit
- fjfkkskdngnf 1mo ago[flagged]
- fridder 1mo agoCopyright is too long (thank you Disney and Pelosi) and I'm all for shortening it. Also I do agree that market dilution is not really defensible. The main reason why models should have been shut down is the sketchy way in which the models were trained.
- snohobro 1mo agoITT: Tell me you don’t have anything worth copyrighting without telling me you don’t have anything worth copyrighting. I worry about some of y’all. After reading the article, I don’t disagree with the EFF. We shouldn’t rewrite copyright law to ban creative generation of {art}. AI generated images only increases competition, which is good for consumers of {art}. I do agree that copyright should still exist to prevent duplicate {art} being sold by a new individual claiming it as their own. I think the tricky thing with AI is that as the end user, it’s still your responsibility to identify when you might be infringing on someone else’s copyright. I imagine that will be very difficult to do. Perhaps we need an AI copyright agent that can ingest your {art} and compare against all existing {art} for percent copyright infringement score. In my opinion, the biggest thing I worry about is small creators having their work stolen, repackaged, and sold by big corporations. Which copyright protects. In fact with AI assisted lawyering, it might actually be even easier for small timers to protect their stuff, just like the big wigs do. Abolishing copyright altogether is just such a wild, knee jerk take.
- LtWorf 1mo ago> it might actually be even easier for small timers to protect their stuff It's easier to not get my stuff stolen if we add superthieves on cocaine?
- teeray 1mo agoI think the asymmetry of fair use is what feels the most wrong about copyright and AI. Forever, fair use has been limited with an iron fist: you can’t rip that DVD because it’s copy-protected; You can only use 10s of that song in your school project. Then these AI companies wield fair use like a sledgehammer in a way that nobody has been able to for decades to enormously profit. Maybe it’s technically correct, but it feels like mass exploitation of a loophole that the legal system has previously taken a dim view towards (see: Aereo).
- LtWorf 1mo agoI'm aware that my GPL3 licensed software has been rewritten using AI to get rid of the GPL license by a large weapons company. Seems to me the EFF is entirely fine with it?
- trashburger 1mo ago20 years of copyright from creation, just like patents. The first two Shrek movies should be public domain by now.
- Planktonne 1mo agoThe EFF has managed to keep to the letter of its principles while discarding the spirit of them; it's now defending enormous commercial concerns that hoover up and hoard data, at the expense of people creating and sharing information. A small number of huge corporations keeping tight control of data and access to data is not a pro-freedom stance.
- UtopiaPunk 1mo agoCopyright is broken and needs major reform, but copyright must exist in some form as long as we are operating under the current economic system. If we want a society where some people have the job of creating art, then they must be compensated for their art. The leverage that artists have to demand compensation is controlling who access to their art. Said another way, the artist owns the right to make copies of their art. That is protected by law. I think copyright as it exists is abused by large companies and stifles culture. Broadly speaking, I think copyright should be a much shorter term, lasting more like 20 years rather than the 100+ years that is typical (life of the author + 70 years). The exact term is debateable, but the purpose should be to allow the artist sufficient time to monetize their creation for themselves, and then allowing the art to belong to the broader culture. There should be a worthwhile financial incentive to create art. My utopian dream would be a radically restructured society where we do not expect people to have "artist" as a career. Instead, people would work 15 hours a week doing the work that must be done to contributue to society. The compensation from that 15 hours of work guarantees all their needs to live in modest comfort. People then have more leisure time to pursue whatever they find self-actualizing, which for many will mean creating art and sharing it with the world.