9 January, 2024

OpenAI Accuses The New York Times Of “Not Telling The Full Story” Over Reproduction Of Its Articles

New York City, USA - May 9, 2013: A The New York Times newspaper in front of The New York Times company office building at 620 8th Avenue, Midtown Manhattan, New York City. It was completed in 2007 and it is owned by The New York Times Company and Forest City Ratner Companies.

OpenAI, the company behind ChatGPT, has said in response to a lawsuit filed by The New York Times alleging copyright infringement that the paper “is not telling the full story”.

At the end of December, the Times filed a lawsuit against OpenAI and Microsoft claiming that the two companies built their AI models by “copying and using millions” of its articles, leading to ChatGPT and Microsoft’s Copilot to “directly compete” with its content.

Microsoft is a major investor in OpenAI and uses its technology as the basis for its Copilot virtual assistant across its programs.

The lawsuit alleges that OpenAI and Microsoft’s large language models (LLMs) “can generate output that recites Times content verbatim, closely summarizes it, and mimics its expressive style.” This “undermine[s] and damage[s]” the Times’ relationship with readers, the outlet alleges, while also depriving it of “subscription, licensing, advertising, and affiliate revenue”.

The lawsuit also alleges that the AI models “threaten high-quality journalism” by damaging the ability of publications to monetise their content.

The Times has not been the only publisher to draw attention to the potential of ChatGPT-like programs to destroy advertising revenue. Nine’s Mike Sneesby has said that the News Media Bargaining Code, which requires tech companies such as Google and Meta to strike commercial deals with online news publishers, should be extended to companies like OpenAI.

News Corp’s Michael Miller also said that “OpenAI has… quickly established a business worth US$30 billion by using the others’ original content and creativity without remuneration and attribution”.

However, in a blog post, OpenAI fired back claiming that the Times wasn’t being completely honest about how its programs reproduce stories verbatim.

“Even when using such prompts, our models don’t typically behave the way The New York Times insinuates, which suggests they either instructed the model to regurgitate or cherry-picked their examples from many attempts,” OpenAI said.

The company also said that it has attempted to limit the amount that its LLMs regurgitate content and that the Times refused to share examples of the reproduction before filing the lawsuit. OpenAI said that any examples of regurgitation “appear to be from year-old articles that have proliferated on multiple third-party websites.” The company did admit that it took down a ChatGPT feature, called Browse, that unintentionally reproduced content.

“Our discussions with The New York Times had appeared to be progressing constructively through our last communication on December 19,” read OpenAI’s blog post.

“The negotiations focused on a high-value partnership around real-time display with attribution in ChatGPT, in which The New York Times would gain a new way to connect with their existing and new readers, and our users would gain access to their reporting. We had explained to The New York Times that, like any single source, their content didn’t meaningfully contribute to the training of our existing models and also wouldn’t be sufficiently impactful for future training. Their lawsuit on December 27—which we learned about by reading The New York Times—came as a surprise and disappointment to us”.

Despite this, the company says that it is still “hopeful for a constructive partnership” with the Times.

In August last year (a year after it released ChatGPT to the public) OpenAI released a new version that allows publishers to block its bots from crawling their site for training data.

Advertise with us

OpenAI The New York Times