A New York Times copyright lawsuit could kill OpenAI::A list of authors and entertainers are also suing the tech company for damages that could total in the billions.
I always say this when this comes up because I really believe it's the right solution - any generative AI built with unlicensed and/or public works should then be free for the public to use.
If they want to charge for access that's fine but they should have to go about securing legal rights first. If that's impossible, they should worry about profits some other way like maybe add-ons such as internet connected AI and so forth.
Not really how it works these days. Look at Uber and Lime/Bird scooters. They basically would just show up to a city and say the hell with the law we are starting our business here. We just call it disruptive technology
Unfortunately true, and the long arm of the law, at least in the business world, isn't really that long. Would love to see some monopoly busting to scare a few of these big companies into shape.
That goes against the fundamental idea of something being unlicensed, meaning there are no repercussions from using the content.
I think what you mean already exists: open source licenses. Some open source licenses stipulate that the material is free, can be modified, etc. and you can do whatever you want with it, but only on the condition that whatever you create is under the same open source license.
Ugh I see what you mean - no I mean unlicensed as in 'they didn't bother to license copyrighted works' and public as in 'stuff they scraped from Reddit, Twitter, and etc. without permission from anyone'.
Would you, after devoting full years of your adult life to the unpaid work of learning the requisite advanced math and computer science needed to develop such a model, like to spend years more of your life to develop a generative AI model without compensation? Within the US, it is legal to use public text for commercial purposes without any need to obtain a permit. Developers of such models deserve to be paid, just like any other workers, and that doesn't happen unless either we make AI a utility (or something similar) and funnel tax dollars into it or the company charges for the product so it can pay its employees.
I wholeheartedly agree that AI shouldn't be trained on copyrighted, private, or any other works outside of the public domain. I think that OpenAI's use of nonpublic material was illegal and unethical, and that they should be legally obligated to scrap their entire model and train another one from legal material. But developers deserve to be paid for their labor and time, and that requires the company that employs them to make money somehow.
Would you, after devoting full years of your adult life to the unpaid work of learning the requisite advanced math and computer science needed to develop such a model, like to spend years more of your life to develop a generative AI model without compensation?
No. I wouldn't want to write a kernel from scratch for free either. But Linus Torvalds did. He even found a way to monetize it without breaking any laws.
If OpenAI owns a Copyright on the output of their LLMs, then I side with the NYT.
If the output is public domain--that is you or I could use it commercially without OpenAI's permission--then I side with OpenAI.
Sort of like how a spell checker works. The dictionary is Copyrighted, the spell check software is Copyrighted, but using it on your document doesn't grant the spell check vendor any Copyright over it.
I think this strikes a reasonable balance between creators' IP rights, AI companies' interest in expansion, and the public interest in having these tools at our disposal. So, in my scheme, either creators get a royalty, or the LLM company doesn't get to Copyright the outputs. I could even see different AI companies going down different paths and offering different kinds of service based on that distinction.
I think it currently resides with the one doing the generation and not openAI itself. Officially it is a bit unclear.
Hopefully, all gens become copyleft just for the fact that ais tend to repeat themselves. Specific faces will pop up quite often in image gen for example.
If LLMs like ChatGPT are allowed to produce non-copyrighted work after being trained on copyrighted work, you can effectively use them to launder copyright, which would be equivalent to abolishing it at the limit.
A much more elegant and equitable solution would be to just abolish copyright outright. It's the natural direction of a country that chooses to invest in LLMs anyways.
The NYT has a market cap of about $8B. MSFT has a market cap of about $3T. MSFT could take a controlling interest in the Times for the change it finds in the couch cushions. I’m betting a good chunk of the c-suites of the interested parties have higher personal net worths than the NYT has in market cap.
I have mixed feelings about how generative models are built and used. I have mixed feelings about IP laws. I think there needs to be a distinction between academic research and for-profit applications. I don’t know how to bring the laws into alignment on all of those things.
But I do know that the interested parties who are developing generative models for commercial use, in addition to making their models available for academics and non-commercial applications, could well afford to properly compensate companies for their training data.
The danger of the rich and evil simply buying out their critics is a genuine risk. After all, it's what happened to Gawker when Peter Thiel decided he personally didn't like them, neutering their entire network.
Regarding OpenAI the corporation, they pulled an incredibly successful bait and switch, pretending first to gather data for educational purposes, and then switching to being a for-profit as soon as it benefited them. In a better world or even a slightly more functional American democracy, their continued existence would be deemed inexcusable.
I completely agree. I don’t want them to buy out the NYT, and I would rather move back to the laws that prevented over-consolidation of the media. I think that Sinclair and the consolidated talk radio networks represent a very real source of danger to democracy. I think we should legally restrict the number of markets a particular broadcast company can be in, and I also believe that we can and should come up with an argument that’s the equivalent of the Fairness Doctrine that doesn’t rest on something as physical and mundane as the public airwaves.
This would bring up the cost of entry for making a model and nothing more. OpenAI will buy the data if they have too and so will google. The money will only go to the owners of the New York Times and its shareholders, none of the journalists who will be let go in the coming years will see a dime.
We must keep the entry into the AI game as low as possible or the only two players will be Microsoft and Google. And as our economy becomes increasingly AI driven, this will cement them owning it.
He's not arguing for OpenAI, but for the rest of us. AI is a public technology, but we're on the verge of losing our ability to participate due to things like this and the megacorps' attempts at regulatory capture. Which they might just get. Their campaign against AI is a lot like governments' attempts to destroy encryption. Support open source development, It's our only chance. Their AI will never work for us. John Carmack put it best.
Fuck "Open"AI, fuck Microsoft. Pragmatism or slavery.
The problem with copyright is that everything is automatically copyrighted. The copyright logo is purely symbolic, at this point. Both sides are technically right, even though the courts have ruled that anything an AI outputs is actually in the public domain.
Works involving the use of AI are copyrightable. Also, the Copyright Office's guidance isn’t law. Their guidance reflects only the office’s interpretation based on its experience, it isn’t binding in the courts or other parties. Guidance from the office is not a substitute for legal advice, and it does not create any rights or obligations for anyone. They are the lowest rung on the ladder for deciding what law means.
Late last year, the New York Times sued OpenAI and Microsoft, alleging that the companies are stealing its copyrighted content to train their large language models and then profiting off of it.
Meanwhile, the Senate Judiciary Subcommittee on Privacy, Technology, and Law held a hearing in which news executives implored lawmakers to force AI companies to pay publishers for using their content.
In its rebuttal, OpenAI said that regurgitation is a “rare bug” that the company is “working to drive to zero.” It also claims that the Times “intentionally manipulated prompts” to get this to happen and “cherry-picked their examples from many attempts.”
A growing list of authors and entertainers have been filing lawsuits since ChatGPT made its splashy debut in the fall of 2022, accusing these companies of copying their works in order to train their models.
Developers have sued OpenAI and Microsoft for allegedly stealing software code, while Getty Images is embroiled in a lawsuit against Stability AI, the makers of image-generating model Stable Diffusion, over its copyrighted photos.
In that 2013 decision, Judge Chin said its technology “advances the progress of the arts and sciences, while maintaining respectful consideration for the rights of authors and other creative individuals, and without adversely impacting the rights of copyright holders.” And a 2023 economics study of the effects of Google Books found that “digitization significantly boosts the demand for physical versions” and “allows independent publishers to introduce new editions for existing books, further increasing sales.” So consider that another point in favor of giving tech platforms room to innovate.
The original article contains 1,628 words, the summary contains 259 words. Saved 84%. I'm a bot and I'm open source!
Hahahahahahaha hahahahahahaha omg, thank you for the very real, actual laugh-out-loud moment.
Now I'm envisioning Picard one one side, BorqBorg (wtf autocorrect?) Queen on the other, and what, Q as judge, looking older by the minute, just hating life.
AI development will not be hamstrung by regulations. If governments want to "regulate" (aka kill) AI, then AI development in their jurisdiction will move elsewhere.