OpenAI calls NYT lawsuit meritless, says it supports journalism

OpenAI responds to New York Times lawsuit, which it says is 'without merit'
NYT alleges copyright infringement, use of articles to train ChatGPT model
Visual artists have long called for tighter regulation over AI-generated art

Updated: Jan 8, 2024 / 11:50 PM CST

(NewsNation) — OpenAI said Monday a lawsuit filed against it by The New York Times is “without merit” and that the company supports journalism and partners with news organization.

The makers of ChatGPT said in a blog post that the Times is “not telling the full story” in its lawsuit that accuses OpenAI and Microsoft of using millions of articles to train their AI tools. OpenAI said it has been working with The Times in the weeks leading up to the lawsuit about a “high-value partnership” regarding attribution within ChatGPT responses.

“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,” the company said.

It called the lawsuit a “surprise and disappointment.”

The newspaper is seeking “billions of dollars in statutory and actual damages” and calls for the companies to destroy any chatbot models or training data that use copyrighted New York Times material.

The lawsuit was filed in Federal District Court in Manhattan and alleges Microsoft and OpenAI used articles without authorization to train chatbots that “now compete with the news outlet as a source of reliable information,” the outlet reported.

In its response letter, OpenAI said the so-called “regurgitation” of content was from years-old content already appearing on other third-party websites.

“It seems they intentionally manipulated prompts, often including lengthy excerpts of articles, in order to get our model to regurgitate,” OpenAI said. “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 doesn’t disclose the source of language that trains its tools but has admitted to using copyrighted works in the past.

Visual artists have long called for tighter regulation over AI-generated art, which is created by models that analyze billions of existing images to create its own.

NewsNation digital reporter Katie Smith contributed to this report.