Meta Platforms (NASDAQ:META) has discussed deals with media publishers to pay to use their content to train its artificial intelligence models, Business Insider reported.
The discussions are said to be at an early stage and could involve partnerships, product and legal teams, the news outlet added. News articles, photos and video content are among the items being discussed internally at Meta.
Meta did not immediately respond to a request for comment from Seeking Alpha.
If Meta were to sign deals with publishers, it would become the latest tech company to do so. OpenAI, the generative AI startup backed by Microsoft (MSFT), has made several deals with media publishers, most recently inking a deal with the Financial Times.
OpenAI signed a deal in December with publishing company Axel Springer — the parent company of Business Insider — to use the media company’s content to train its AI models.
It also recently signed deals with French newspaper Le Monde and Spanish media conglomerate Prisa Media and has existing deals with the Associated Press, the American Journalism Project and NYU.
Publishers including News Corp. (NWS) (NWSA), Gannett (GCI) and others have also reportedly held talks with OpenAI to license their content.
The New York Times Company (NYT) sued Microsoft and OpenAI for copyright infringement in late December, alleging the tech companies illegally used the newspaper’s content to train artificial intelligence models.
Google (GOOG) (GOOGL) inked a deal with Reddit (RDDT) earlier this year to use Reddit’s content to train its AI models.
Media reports last month said that Google had signed a deal with News Corp. to develop AI-related content and products. However, a News Corp. spokesman told Seeking Alpha that the company “absolutely” did not have an AI content licensing deal with Google, but it did have a “number of partnerships with Google across our businesses.”
Apple (AAPL) has also reportedly held talks with several publishers about using their content to train its respective AI models.