A recent setback has occurred for OpenAI in its copyright dispute with The New York Times and other news organizations. A federal judge in Manhattan ruled that OpenAI must provide millions of anonymous chat records from ChatGPT users. This decision, issued by U.S. District Judge Ona Wang this Wednesday, found that these 20 million chat records are crucial to the news organizations' claims, and disclosing them would not infringe on user privacy.
Judge Wang rejected OpenAI's objections regarding privacy, stating that due to the highly sensitive and private nature of the content involved, multiple protective measures have been taken. An OpenAI spokesperson expressed concern, citing a blog post by the company's Chief Information Security Officer, Dane Stuckey, which stated that The New York Times' request to disclose chat records "ignores long-standing privacy protections" and "deviates from common security practices."
Meanwhile, OpenAI has appealed to the lead judge in the case, Sidney Stein. The lawsuit was also filed by some newspapers owned by MediaNews Group, a subsidiary of Alden Global Capital. Frank Pine, executive editor of MediaNews Group, stated in a statement that OpenAI's management "is delusional about hiding how their business model relies on stealing the work of diligent journalists."
This case originated from The New York Times' lawsuit in 2023, involving multiple lawsuits by copyright holders against tech companies including OpenAI, Microsoft, and Meta, as they used their materials without authorization to train artificial intelligence systems. News organizations argued that chat records are essential evidence to determine whether ChatGPT copied their copyrighted content and to counter OpenAI's claim that they "hacked" the chatbot to create evidence.
OpenAI argued that submitting these chat records would expose confidential user information and claimed that "99.99%" of the records are irrelevant to the infringement allegations. Judge Wang emphasized in her initial order that user privacy would be protected through the company's "comprehensive de-identification" and other safeguards, and reiterated that these measures would "reasonably alleviate related privacy concerns." She required OpenAI to submit the chat records within seven days after removing user identifying information.
Key Points:
📰 OpenAI was ordered by a judge to disclose 20 million anonymous ChatGPT user chat records as evidence in a copyright dispute.
🔒 The judge stated that disclosing chat records would not infringe on user privacy, as multiple protective measures have been taken.
📜 The case was initiated by The New York Times, involving multiple news organizations, aiming to confirm whether ChatGPT infringes on copyrighted content.
