OpenAI is in the middle of a legal battle in India after ANI, a leading news agency, accused the company of copyright infringement. ANI claims that OpenAI used its articles to train ChatGPT without permission and is asking the company to delete the data. However, OpenAI has stated that it can’t comply with this request due to legal requirements in the United States, where it is bound to keep the training data intact.
OpenAI explained in a court filing that deleting ANI’s data would go against US laws, particularly because of ongoing copyright cases, including one with The New York Times. The company also argued that Indian courts don’t have the jurisdiction to enforce the request, as OpenAI doesn’t have any offices or servers in India.
ANI’s lawsuit claims that ChatGPT has been generating parts of its articles in response to user queries, which they believe harms their business and creates unfair competition. The news agency also pointed out that OpenAI has partnerships with major global media outlets like Time Magazine and The Financial Times, making the playing field uneven for local publishers like ANI.
In response, OpenAI denied any wrongdoing, saying that ANI had used its own articles as prompts to manipulate ChatGPT’s outputs. This exchange of accusations is a sign of the rising tension between AI companies and content creators, as more and more creators question how their content is being used by AI models.
OpenAI’s defence focuses on its operations in the US. The company argues that it’s legally required to maintain its training data under US law and that selectively deleting data, as ANI demands, would violate these regulations. Additionally, OpenAI highlighted that it doesn’t operate in India and has no physical presence or servers there, so Indian courts shouldn’t have authority over the matter.
This situation raises interesting questions about how countries can regulate foreign companies operating in their territories, especially when it comes to something as complicated as AI training.
This case is just one example of a larger trend of copyright disputes between content creators and AI companies. As AI models like ChatGPT are trained on vast amounts of publicly available data, concerns about copyright infringement and fair use are becoming more common. OpenAI has started to sign licensing agreements with some media companies, but the bigger issues of compensation and consent are still up in the air.
With the case set to be heard by the Delhi High Court on 28 January, the outcome could have a big impact on how India handles copyright issues related to AI. It might also set the stage for how other countries deal with similar disputes in the future.
Link to article –
Cannot delete training data as Indian courts have no jurisdiction: OpenAI in copyright case
