US AI giant OpenAI has told an Indian court that Indian judiciary does not have jurisdiction over a copyright breach case brought by a local news agency, arguing it had legal obligations in the US. Last year, India’s ANI agency sued OpenAI in Delhi High Court, accusing the AI firm of using its content to train its AI models without authorisation. The agency sought $231,000 in damages from the firm and also assurance that it will delete all of its content from its data centres.

Advertisment

Now, OpenAI has told the court that any order in India to remove training data used to train ChatGPT would go against its legal obligations, news agency Reuters reported. In November, OpenAI also assured the court that it won’t use ANI’s content anymore.

The company questioned Indian court’s jurisdiction over the matter, citing a lack of official presence in the country.

OpenAI is facing copyright breach lawsuits across the globe, with several news agencies dragging the AI giant to courts seeking compensation for their content ‘unlawfully used’ to train AI models. On the contrary, OpenAI maintains it uses publicly available data.

Advertisment

Also read: Musk ridicules funding viability of Trump's $500 billion AI project Stargate: 'Actually wasn't there'

The matter is listed for hearing at the Delhi High Court on January 28.

In the US, The New York Times, The New York Daily News, and the Center for Investigative Reporting are leading the legal campaign against OpenAI and its backer Microsoft.

Advertisment

Also read: China considers selling TikTok US to Elon Musk amid ban threat, ByteDance denies rumours

The news organisations argue that ChatGPT was fed with millions of copyrighted works from their websites, including archives, for training purposes, and it was done without consent or remuneration.

Earlier, other news publishers, including the Associated Press, News Corp., and Vox Media, struck a deal with OpenAI over content sharing. But the Times and company are going full offensive, who argue that ChatGPT's dataset must be deleted and billions should be paid in damages.

(With inputs from agencies)