New York Times sues OpenAI and Microsoft, over copyright violations
The New York Times has taken legal action against OpenAI and Microsoft in a United States court, asserting that these companies utilised millions of articles from the publication without authorisation in training their powerful AI models.
In a lawsuit filed on Wednesday, The Times claimed that both companies employed its journalism in their AI chatbots without consent or compensation, essentially capitalising on the extensive investments made by The Times in producing high-quality content.
The issue of copyright has become a major point of contention in the rapidly growing generative AI sector, with creators such as publishers, musicians, and artists increasingly resorting to legal measures to ensure fair compensation for their content being used in technological developments.
In response to the burgeoning prominence of AI chatbots, The New York Times opted for a confrontational stance by suing, differing from other media entities that have engaged in content agreements with OpenAI, such as Germany’s Axel Springer or the Associated Press.
The lawsuit highlighted the criticality of safeguarding independent journalism, stressing that if news organizations like The Times cannot protect their work, the consequences for society would be substantial, leading to a reduction in journalistic output with far-reaching implications.
The legal action seeks damages and demands an order for both companies to cease using The Times’ content for AI model training, along with the deletion of previously acquired data. While the exact figure for damages remains unspecified, The Times suggested potential losses amounting to billions of dollars.
Despite attempts to negotiate a content agreement, OpenAI and Microsoft defended their use of the content, claiming it fell under the category of “transformative” technology, thereby implying it did not necessitate a commercial arrangement. The lawsuit refuted this argument, stressing that using The Times’ content without compensation to create competing products was unjustifiable.
Moreover, the lawsuit alleged that the AI-generated content closely imitated the style of The New York Times, occasionally attributing false information to the reputable news source, while also underscoring the significance of decades-worth of archived news in training AI models.