OpenAI Rejects Claim: Did Not Use Indian Media Content To Train ChatGPT

India-West News Desk

NEW DELHI – OpenAI is opposing efforts by Indian media groups, including those linked to Gautam Adani and Mukesh Ambani, to join a copyright lawsuit against the U.S. company, asserting that it has not used their content to train ChatGPT, a legal filing reviewed by Reuters shows.

The Microsoft-backed firm stated it is not required to form partnerships with media organizations to access publicly available content, according to its 31-page court filing, reported by Reuters for the first time. The lawsuit originated last year when Indian news agency ANI accused OpenAI of using its published content without permission for AI training. Since then, book publishers and media houses have sought to join the case.

Among them are Adani’s NDTV, The Indian Express, The Hindustan Times, and the Digital News Publishers Association, which represents several outlets, including Ambani’s Network18. They allege OpenAI has scraped their news websites and reproduced content on ChatGPT. However, in its February 11 filing, OpenAI denied using any of their material to train its AI models, Reuters said.

While OpenAI has licensing agreements with several news publishers globally, Indian media groups argue the company has not made similar arrangements in India. OpenAI clarified in its filing that its global partnerships do not involve licensing content for AI training and that using publicly available content is permissible under Indian copyright law.