OpenAI filed a motion be dismissed on Monday new york times” Generative AI company ChatGPT has filed a lawsuit over claims of copyright infringement.
in this case, times It claimed that its flagship AI bot was replicating newspaper reports and spreading hallucinatory content attributed to the newspapers. times It damaged the company’s reputation.
OpenAI’s motion also states: times They “hacked” ChatGPT with brute force and “regurgitated” training data into the chatbot.
“The truth that will be revealed in the course of this case is times “We paid someone to hack OpenAI’s products,” the technology company’s lawyers argued in the filing. He said it “took tens of thousands of attempts to produce the highly anomalous results that constitute Indictment J” and called this “a long-term, large-scale effort to hack OpenAI’s models.” called it
According to OpenAI, the results are: Showcasedetails a scene where ChatGPT spits out a passage verbatim. new york times Articles can only be generated by a company’s repeated attempts to generate them when prompted. “Ordinary people would not use OpenAI’s products in this way,” the company’s lawyers argued.
In its initial response to the lawsuit on OpenAI’s blog, it said: Had made In the same argument, “Even when using such prompts, our model typically new york times “This suggests that they either instructed the model to regurgitate it or that they cherry-picked examples from many attempts,” as well as the lawsuit. times Developing a basis for copyright claims required repeated efforts.
The company’s lawyers also challenged the company’s use of a large corpus of Internet data to train its products. new york timesmay constitute copyright infringement.
OpenAI argues in its application that reproducing excerpts from copyrighted text is fair use; times did just that with a report using block quotes.
They also cite decisions in several other lawsuits filed against generative AI companies in the past year as precedents for this argument, including a lawsuit filed by the authors against OpenAI. (They are currently trying to consolidate their case and litigation) times), the other comes from an artist who opposes Midjourney, whose products generate images.
According to these rulings, OpenAI’s lawyers say the fact is that ChatGPT does not generate the full text of the article, but provides the source of the article. times It means that in its output it does not infringe copyright in the way that the newspaper is accusing it of.
on the other hand, times It claims that with AI responses, readers no longer need to visit a website to get information. The company says this is especially harmful for promotional articles it publishes to endorse its products. wire cutter This section contains affiliate links.
However, OpenAI rejected the idea that it would cause harm in this way, stating that the example: times What is offered is that ChatGPT responds by “directing the user to Wirecutter itself and providing a non-verbatim summary of the recommendations.”
Along with the motion to dismiss the lawsuit, OpenAI also requested that oral arguments be held to discuss the dismissal.
“ChatGPT never new york times” wrote OpenAI’s lawyers. “No one in the real world would use ChatGPT or any of his other OpenAI products for such purposes.”
We crawl the web for you so you don’t have to.
Sign up for the Daily Dot newsletter to get the best and worst of the internet delivered to your inbox every day.
*First published: February 27, 2024 at 4:39pm CST
Marlon Ettinger
Marlon Ettinger writes about politics, crime, and culture. For tips and ideas please email him at marlonjettinger@gmail.com.