OpenAI, owner of the highly successful artificial intelligence (AI) chatbot ChatGPT, has opposed the lawsuit filed against it by US publication, The New York Times. The suit accuses the AI firm of violating copyright laws by deploying a large volume of Times’ journalistic content to train its systems.
In a statement released last week, OpenAI claims that it supports journalism and partners with news organisations. The company says the Times’ lawsuit accusing it of copyright infringement is “without merit”, and that it disagrees with the claims in the lawsuit brought by the publication.
The case, which was filed in December 2023 against OpenAI and its major investor Microsoft, seeks billions of dollars in damages, claiming OpenAI’s chatbot competes with the Times as a “reliable” source of information. The Times says OpenAI has illegally used millions of its articles to train its systems for information that will be publicly available.
OpenAI, on the other hand, says it collaborates with news outlets, creating new opportunities. The firm claims the Times is “not telling the full story”. “We work hard in our technology design process to support news organizations,” the ChatGPT owner says.
“We’ve met with dozens, as well as leading industry organizations like the News/Media Alliance, to explore opportunities, discuss their concerns, and provide solutions.”
OpenAI says it seeks to support journalism by benefiting reporters and editors through its products, which can assist them with intensive tasks, including analysing large public records and translating news stories.
The firm also claims to teach its AI models about the world by training them on “historical” content that is not publicly available. Moreover, it will provide additional ways for news publishers to connect with readers by displaying real-time content with attribution in ChatGPT.
“We regard The New York Times’ lawsuit to be without merit,” OpenAI adds. “Still, we are hopeful for a constructive partnership with The New York Times and respect its long history, which includes reporting the first working neural network over 60 years ago and championing First Amendment freedoms.”
Since the rise of ChatGPT in October 2022, the debate surrounding the use of copyrighted material has intensified. AI firms argue that training their language models on available content is highly beneficial for the public, while content creators and publishers strongly oppose the rapid developments. They cite ethical concerns and level accusations of unauthorised use of copyright-protected content against AI companies.
