"ChatGPT's Unauthorized Use of Articles"... US NYT Sues OpenAI and MS

"Billions of Dollars in Losses... Fare Evasion"
'Game of Thrones' and Getty Images Sue AI

The New York Times (NYT) in the United States has filed a lawsuit against OpenAI, the developer of 'ChatGPT,' and Microsoft (MS), alleging copyright infringement of its publications.

On the 27th (local time), NYT claimed in a complaint submitted to the U.S. District Court for the Southern District of New York that millions of articles it produced were used without authorization to train generative artificial intelligence (AI), and that ChatGPT is competing with NYT as a reliable information provider. This has caused legal and actual damages amounting to billions of dollars due to reduced advertising, licensing, and subscription revenues. Although NYT did not disclose the specific amount of the lawsuit, it demanded damages from OpenAI and MS and the destruction of its data used for AI training.

Prior to this lawsuit, NYT reportedly engaged in negotiations with OpenAI and MS regarding news usage fees in April, but the talks broke down, leading to legal action.

In the complaint, NYT stated, "OpenAI and MS are trying to free-ride on our substantial investment in journalism," and "They use our content without paying for it, replacing us and stealing our readers." Users satisfied with ChatGPT's answers do not visit the NYT website, resulting in decreased advertising and subscription revenue, and ChatGPT provides excerpts from articles that require paid subscriptions as-is. There are also concerns that the 'AI hallucination' phenomenon, where incorrect information is input and provided as is, could damage the value of NYT's brand.

NYT also argued, "As readers move online, dozens of newspapers and magazines are struggling," while "AI companies like OpenAI, which use various online documents such as news, poems, and scripts for chatbot training, are raising billions of dollars in funding." NYT expects this lawsuit to test the legal limits of generative AI technology and have a significant impact on the media industry.

Currently, many media companies point out that generative AI companies use their news content without authorization for AI training. On the other hand, AI companies have argued that using publicly available content from the open internet for AI training is permitted under the U.S. copyright law doctrine of 'fair use.' Fair use means that in non-commercial cases, copyrighted material can be used without permission within limits that do not unfairly harm the copyright holder's interests. However, NYT holds the position that this provision should not apply because AI can provide its articles verbatim.

This is not the first time legal battles have arisen over copyright infringement by generative AI. Previously, cultural figures including George R.R. Martin, the original author of the popular TV drama 'Game of Thrones,' as well as famous writers and comedians, filed lawsuits against OpenAI. In February, Getty Images sued the UK-based image generation AI company StabilityAI for unauthorized use of images.

In response, OpenAI has signed news usage agreements with some media companies. In July, OpenAI signed a contract with the Associated Press (AP). In April, it entered into a three-year news usage agreement with media group Axel Springer, which owns Politico, Business Insider, and Bild.

This content was produced with the assistance of AI translation services.