The Delhi High Court has issued a significant ruling in ANI Media Pvt. Ltd.'s copyright infringement suit against OpenAI, one of the first Indian cases to squarely address whether using copyrighted news content to train a large language model, and whether an AI chatbot's outputs, can attract copyright liability and the fair dealing exception under the Copyright Act, 1957.
ANI's case centres on OpenAI's alleged use of ANI's news content, without licence, to train ChatGPT, and on ChatGPT's ability to reproduce or closely paraphrase that content in response to user queries — including, ANI alleged, misattributing fabricated content to ANI. OpenAI's defence has focused on the servers used for training being located outside India, and on the fair-dealing and research-related exceptions available under Indian copyright law.
As one of the first substantial Indian judicial engagements with generative AI and copyright, the case is being closely watched by news publishers, AI developers and technology companies operating in or serving the Indian market, since its eventual resolution will materially shape how Indian copyright law applies to AI training data and model outputs.