Man reaches a $ 1.5 billion settlement with authors in the case of copyright for the teacher

AI Startup approved two authors for about $ 3,000 per book of about 500,000 works, after accusing it of downloading millions of pirated texts from Shadow libraries to train the Great Language Model, Claude. As part of the deal, Anthropor will also destroy the data accused of illegally acquired.

AI announced rapidly growing growth earlier this week that it just raised an additional $ 13 billion in new investment capital financing in a deal estimated at $ 183 billion. I also mentioned that he is currently going to generate at least $ 5 billion in revenue over the next 12 months. The settlement will reach nearly a third of this number or more than the new tenth of the new financing it has just received.

Although the settlement does not specify a legal precedent, experts said it will probably work as an anchor figure for the amount that other major artificial intelligence companies will need to pay if they hope to settle similar copyright -violating cases. For example, a number of authors sue Dead To use their books without permission. As part of this lawsuit, Meta has been forced to unveil the company’s internal emails that indicate that it has intentionally used a library of pirated books called Libgen – one of the same libraries that a person used. Openai and its partner Microsoft They also face a number of copyright violations, including those presented by the author guild.

Aparna Sridhar, Deputy General Adviser to Anthrop, said. luck In a statement: “In June, the District Court issued a prominent ruling on the AI ​​Law on Artificial Intelligence and Publishing Rights, and found that the Antarubor approach to training artificial intelligence models constitutes fair use. If today, if approved, they will solve the remaining prosecutors.

A lawyer for authors who filed a lawsuit against Antarubor said that the settlement would have long -term traces.
“This historical settlement greatly exceeds any other well -known restoration of copyright. It is the first of its kind in the era of artificial intelligence. It will provide meaningful compensation for each category work and put a precedent that requires artificial intelligence companies to pay copyrights.” Partz and others. againstHe said in a statement. “This settlement sends a strong message to both companies and Amnesty International that taking copyrights from these pirate sites is wrong.”

The case, which was originally appointed to the trial in December, had been subjected to anthropologist, to damage to 1 trillion dollars if the court found that the company intentionally violated the copyright law. Santa Clara Ed Li Law said that if the Antarbur loses the trial, it may have “at least the possibility of the end -of -business obligations.” Antarubor mainly agreed to the conclusion of me, as she wrote in a court file that she felt “excessive pressure” to settle the case given the volume of possible damage.

Anthropor faced the danger facing the means that I used to obtain books protected by copyright, rather than the fact that they used books to train artificial intelligence without explicit permission from copyright. In July, the American Partial Court judge, William AlsUP, spent the use of copyright books to create a model of artificial intelligence that constitutes a “fair use” that no specific license was needed. But Alsup and then focused on the claim that Antarbur has used digital libraries for pirated books for some of the data that I fed at least the artificial intelligence models, instead of buying copies of books legally. The judge suggested in a decision allowing the case to trial that he was tending to consider this a violation of copyright, regardless of what a person did with pirated libraries.

By settling the case, man has avoided an existential risk of her work. However, the settlement is much higher than some legal experts who were expecting. The proposal is now seeking to obtain initial approval on what is alleged to be “the greatest recovery of copyright and reported in history.”

James Gremilman, a law professor at Cornell Law College and Cornell Tech, described it as a “modest settlement”.

“Not trying to solve all copyright problems about AI Tawalidi. Instead, he focuses on what Judge Alsup is so terrible that the Antarbur has done so: download books in large quantities of shadow libraries instead of purchasing and wiping copies. luck.

He said that the settlement helps in proving that artificial intelligence companies need to obtain their training data legally, but they do not answer other copyright questions facing artificial intelligence companies, such as what they have to do to prevent the models of artificial intelligence from producing outputs that violate copyright. In many cases, they are still pending against artificial intelligence companies – including the case New York Times It was advanced against Openai and a case presented by Warner Brothers in the Film Studio this week against Midjourney, a company that manufactures Amnesty International that can create photos and videos – copyright holders claim that artificial intelligence models produce identical outputs or are very similar to copyright -producing work.

“Warner Bruce’s recent lawsuit against Meadjurny focuses on how to use Midjournyy to produce pictures of super drivers and other copyrights protected.”

While legal experts say the amount can be controlled for a human -size company, Luke McDonalg, assistant professor of law at LSE, said that the issue may have an impact in the direction of the river course on the smaller artificial intelligence companies if a precedent of action for similar demands was set.

“No. $ 1.5 billion, as the total amount of settlement, indicates the type of level that can solve some other publishing rights from artificial intelligence. It can also indicate the road to licensing copyright to train artificial intelligence.” luck. “This type of amount – $ 3000 per work – can be controlled for a high -level company of artificial intelligence companies and others. It may be less than that for smaller companies. ”

A precedent of work for other artificial intelligence companies

The settlement was the moment “NAPSter to iTunes” for Amnesty International.

“This settlement indicates the beginning of the necessary development towards a legal licensing system based on training data,” she said.. She added that the settlement can distinguish “the beginning of a more mature and sustainable environmental system where creators are compensated, such as how to adapt the music industry to digital distribution.”

Ziniti also pointed out that the volume of settlement may force the rest of the industry to obtain more serious in licensing copyright works.

She said: “The argument that it is very difficult to track and push the training data is a red herring because we have enough deals at this stage to show that this can be done,” noting that news publications, including Axel Springer and Vox, entered with Openai. “This settlement will push other artificial intelligence companies to the negotiating table and accelerate the creation of a real data market, which is likely to include API authentication and revenue sharing models.”

Fortune Global Forum Returns 26 to 27 October, 2025 in Rydah. Executive chiefs and world leaders will meet for a dynamic event for the call only forms the future of business. Apply for an invitation.