In its lawsuit, filed in the U.S. District Court for the Southern District of New York, CIR claims that OpenAI “copied, used, summarized and displayed CIR's valuable content without CIR's permission or approval, and without any compensation to CIR.”
Since it was made publicly available in late 2022, OpenAI's ChatGPT chatbot has been crawling the web to provide answers to users' questions, often relying heavily on copy taken directly from news articles.
“When Defendants introduced works of journalism into their training set, Defendants had a choice: either to respect the works of journalism or not,” the plaintiffs wrote in the complaint. “Defendants chose the latter.”
In a press release Thursday, the nonprofit's CEO, Monica Bauerlein, accused the defendants of “free-riding behavior.”
“OpenAI and Microsoft began collecting our articles to make their products more powerful, but unlike other organizations that have licensed our material, they never asked for permission or offered compensation,” Bauerlein said.
CIR, which runs Mother Jones and the audio show Reveal, also alleges in its lawsuit that OpenAI “trained ChatGPT not to acknowledge or respect copyright, and they did this all without permission.”
Citing the Digital Millennium Copyright Act, the group said it was seeking “actual damages and defendants' profits, or statutory damages of not less than $750 for each infringing work and not less than $2,500 for each DMCA violation.”
OpenAI and Microsoft did not immediately respond to requests for comment.
The news industry as a whole is struggling to maintain enough advertising and subscription revenue to fund costly reporting operations, and many publications are aggressively trying to protect their businesses as AI-generated content becomes more prevalent.
In December, The New York Times filed suit against Microsoft and OpenAI, alleging intellectual property infringement related to the company's journalistic content that appeared in ChatGPT's training data. According to documents filed in the U.S. District Court for the Southern District of New York, the Times said it seeks to hold Microsoft and OpenAI liable for “billions of dollars in combined statutory and actual damages” related to the “unlawful copying and use of The Times' uniquely valuable work.” OpenAI disagreed with The Times' description of the incident.
The Chicago Tribune, along with seven other newspapers, filed a similar lawsuit in April.
Outside of the news, a group of prominent US authors including Jonathan Franzen, John Grisham, George R.R. Martin and Jodi Picoult sued OpenAI last year, alleging copyright infringement when their work was used to train ChatGPT.
But not all news organizations are up for the fight, and some are teaming up with OpenAI: On Thursday, OpenAI and Time announced a “multi-year content deal” that will give OpenAI access to current and archived articles from Time's more than 100-year history.
According to the press release, OpenAI will be able to display Time content within its ChatGPT chatbot to respond to user questions, and will be able to use Time content “to enhance its own products,” or perhaps to train its artificial intelligence models.
OpenAI announced a similar partnership with News Corp. in May, giving OpenAI access to current and archived articles from publications including The Wall Street Journal, MarketWatch, Barron's and the New York Post. Reddit also announced a partnership with OpenAI in May to allow it to train AI models on Reddit content.
clock: Microsoft puts itself at a disadvantage in AI due to Apple and OpenAI deal
