OpenAI Responds to New York Times Lawsuit Over ChatGPT’s Use of News Articles
OpenAI, the company behind the popular AI tool ChatGPT, has issued a response to the legal action taken against it by The New York Times. In a lawsuit filed in late 2023, the US media outlet alleged that AI chatbots like ChatGPT are trained on millions of articles published by The New York Times, resulting in competition between the newspaper and these chatbots as a source of reliable information.
Disputing the claims made by The New York Times, OpenAI stated that the newspaper is not sharing the full story. The AI company emphasized that it collaborates with news organizations and provides an opt-out option for training data. OpenAI argued that training AI models using publicly available internet materials is fair use, as it aligns with long-standing and widely accepted precedents.
OpenAI further explained that being good citizens is more important to them than asserting legal rights. The company took the lead in offering a simple opt-out process for publishers to prevent their tools from accessing the publishers’ sites. The New York Times adopted this opt-out process in August 2023.
Addressing concerns of regurgitation or memorization, OpenAI assured that such failures in the AI’s learning process are rare. To prevent these instances, OpenAI has implemented measures that limit inadvertent memorization and regurgitation in the model’s outputs. The company also expects its users to act responsibly and not manipulate the models to regurgitate content, which goes against its terms of use.
OpenAI underscored that the model’s learning process relies on an enormous aggregate of human knowledge, meaning any single sector, including news, represents only a small fraction of the overall training data. OpenAI disputed the significance of any single data source, including The New York Times, in the intended learning of their model.
In response to The New York Times’ lawsuit, OpenAI expressed surprise and disappointment, as they had engaged in discussions with the media outlet about potential collaboration for several months. OpenAI claimed that despite The New York Times mentioning instances of regurgitation, they refused to provide any examples, despite OpenAI’s commitment to investigate and rectify any issues.
OpenAI believes The New York Times’ lawsuit lacks merit but remains hopeful for a constructive partnership with the media outlet, acknowledging its long history. However, copyright concerns extend beyond this lawsuit, with a report from the News Media Alliance highlighting the use of copyrighted content in training datasets by large language models.
Authors and content creators have also raised concerns about the unauthorized use of copyrighted materials by these platforms. The legal battle between OpenAI and The New York Times is likely to shed light on the complex issues surrounding AI, news organizations, and intellectual property rights.
The outcome of this lawsuit could have significant implications for the future of AI technology, fair use policies, and the relationship between AI models and the news media. Organizations and individuals in the technology and journalism sectors will be closely monitoring the developments in this case.
