Overview
- The court dismissed ANI’s request for a temporary ban, finding on a prima facie basis that OpenAI’s storage and use of ANI articles for model training does not amount to copyright infringement.
- Justice Amit Bansal said in a July 24 interim order that using ANI’s published reports to train large language models prima facie fits the fair‑dealing exception for research and private use under Section 52(1)(a) of the Copyright Act.
- The judge found that ChatGPT outputs, including those produced with retrieval‑augmented generation, were not substantially similar to ANI’s works and that ANI had not shown the models memorised or regurgitated its reports.
- The court refused an injunction because it said an interim ban would cause irreparable harm to OpenAI and the public interest, noting publishers can technically block crawlers if they do not want content used.
- The substantive suit remains in the Delhi High Court, with amici and competing publisher interventions in place, and parallel cases and discovery fights in other countries likely to shape evidence, remedies and possible licensing outcomes.