Intellectual Property: AI wins creators’ expenses

AI companies are winning the fight for copyright, says Nitish Pahwa in Slate. In recent weeks, two different federal judges in the Northern District of California have issued legal rulings, “a largely exempts meta and humanity, saying that they will use published books for training generative bots.” The AI company fired to hoover data from every corner of the internet, including the intellectual property of successful authors such as Ta-Nehisi Coates and Richard Kadrey. However, the two judges suggest that this is all fair use and therefore “completely legal.” There are many copyright lawsuits over AI companies still passing the legal system, but these decisions may set a disastrous precedent at the moment “the moment in which AI is overturning the already long-besieged creative sector.”

Silicon Valley is blessing, but the devil is in detail, Addie Robertson said. As to how closely the final results resemble copyrighted media, the ruling that “training or media was fed to the model and did not reach the question” “concerns “particularly deal with training.” It’s “very appropriate” and is linchpin in the legal battle between Open Eye and the New York Times, claiming that “chatgpt can circulate most of the verbatim narrative.” In his ruling, Judge William Alsup also criticized humanity for downloading more than seven million books, “copyright infringement of scale,” from the online “shadow library.” US District Judge Vince Chhabria added another wrinkle. He awarded in Meta’s favor, but “provided the plaintiffs a roadmap to win their claims in a similar case.” He “almost begged” for artists and authors to present him with evidence of how the AI-generated slops that are flooded in the market were affecting the sale of the original Creative Works.

Chatbots exist only “because of other people’s stiff and uncompensated work,” said Dave Lee of Bloomberg. They are not “creative” themselves. AI “doesn’t generate new values as it would transfer. One place, the original source, to another source: itself.” Regardless of how similar the end output is to the original work the model devoured, AI’s ease of “is a shortcut that reduces the need to visit a library, subscribe to a newspaper, or buy books. If the stated goal of the copyright law is to advance the original work in the original text, AI companies must pay the sender. The tech industry may have determined that the authors “should put aside to serve the great calling of AI,” but lawmakers don’t need to.