Delhi High Court: OpenAI training on ANI news was not copyright infringement
India’s Delhi High Court ruled that OpenAI’s use of ANI news content to train ChatGPT did not infringe copyright because ANI did not show ChatGPT reproduced its specific reports. The decision turns on distinguishing training on copyrighted material from outputting protected expression.
Key takeaway
For AI builders and rights holders in India, the practical fight may shift from proving data was used in training toward proving specific expressive copying in outputs.
Context
Techmeme summarized a Delhi High Court ruling that OpenAI’s use of ANI content to train ChatGPT was not copyright infringement. The court found ANI failed to demonstrate that ChatGPT reproduced its specific reports.
That framing treats training ingestion and output similarity as legally distinct questions. If that standard holds, plaintiffs may need concrete evidence of reproduced expression rather than showing only that protected works entered a training set.