A group of 17 news organizations that are suing OpenAI for its use of their content to train artificial intelligence models asked a federal court for sanctions because they allege the company lied about its ability to search its own datasets. File Photo by Adam Vaughan/EPA
July 9 (UPI) — More than a dozen news organizations asked a court to sanction OpenAI for withholding evidence in lawsuits filed against the company for copyright infringement.
The New York Times, New York Daily News, The Intercept and 14 other news organizations asked a federal court on Thursday for sanctions against the artificial intelligence company for lying about its ability to provide data showing how it has used copyrighted material to train its models.
The companies had sued OpenAI for violating copyright law by using their content to create a secondary product — its AI models — without paying for it, The New York Times and Variety reported.
In court, the AI company had said it could not search training datasets and output data, but earlier this year one of the company’s employees said during a deposition that the data could be accessed.
“The evidence is in OpenAI’s training data sets and ChatGPT output logs,” the organizations said in the court filing.
“But instead of just producing that evidence at the start of the case and focusing on the merits of its fair use defense, OpenAI chose obstruction,” they said.
In addition to accusing OpenAI of lying about searching for the organizations’ content in its data, they allege that the company deleted data logs, which would violate a court order to preserve relevant evidence.
An attorney for the organizations said in a statement that OpenAI had claimed that searching its ChatGPT outputs was “infeasible, burdensome and invasive of users’ privacy” but then lied about having already done searches.
OpenAI called the news organizations’ allegations in the filing “blatantly false,” and said that its use of their content falls under “the long-established principles of fair use.”