ANI appeals Delhi High Court ruling on AI training with news content

Indian news agency ANI has filed an appeal against a Delhi High Court order that permitted OpenAI to train its ChatGPT model using ANI's journalistic content. The ruling, which classified AI model training as "fair dealing" under copyright law, is now set to be reviewed by a higher bench.

In July 2026, a single judge at the Delhi High Court denied ANI's request for an interim injunction that would have prevented OpenAI from using its content. The judge reasoned that the process of training large language models (LLMs) could be considered a form of research, exempt from copyright infringement under Section 52(1)(a) of India's Copyright Act, 1957. The court also found that ChatGPT's outputs did not substantially reproduce ANI's original works and that ANI had failed to demonstrate that the models memorized or regurgitated its content.

The appeal carries significant weight as it challenges the first Indian judicial interpretation on the relationship between AI training and copyright. This interim order has also served as a quiet benchmark for other publishers in licensing negotiations. ANI's appeal specifically contests whether commercial AI model training qualifies as "fair dealing," especially when its purpose is commercial.

ANI initiated its lawsuit against OpenAI in November 2024. The current appeal pertains only to the refusal of interim relief, while the underlying lawsuit remains pending. ANI's specific grounds for appeal are not yet public but are expected to address the application of the research exception to commercial activities and evidence regarding the models' potential memorization capabilities.

The case holds broad implications for the AI industry and publishers' rights, as it will shape future practices concerning the use of copyrighted material for AI development in India and potentially beyond.