The Delhi High Court has ruled that OpenAI’s use of ANI’s content to train ChatGPT does not amount to copyright infringement. Here’s what the landmark judgment means for AI and copyright law in India.
Delhi High Court Rules in Favour of OpenAI
The Delhi High Court has ruled that OpenAI’s use of news agency ANI’s content to train ChatGPT does not constitute copyright infringement under Indian law. The decision marks the first substantive judicial finding in Indiaon whether artificial intelligence companies can use copyrighted news content to train large language models without obtaining a licence.
The ruling is significant because similar legal questions are currently being examined in lawsuits against OpenAI in the United States and Canada.
Why Did ANI Sue OpenAI?
The case began in November 2024, when Asian News International (ANI) sued OpenAI in the Delhi High Court. ANI alleged that the company had used its published news content without permission to train its AI models and had not obtained a licence for doing so. The news agency also claimed that ChatGPT generated responses attributing fabricated stories to ANI, raising concerns over misinformation and the use of its brand.
What the Court Said
Justice Amit Bansal ruled in favour of OpenAI, stating that ANI had failed to demonstrate that ChatGPT had memorised or reproduced its copyrighted news reports in responses generated for users. According to the court, there was no evidence showing that ChatGPT reproduced ANI’s original articles in a manner that amounted to copyright infringement.
The court further held that OpenAI’s storage and use of ANI’s news articles for training the models underlying ChatGPT is protected under the fair dealing exemption for research provided in India’s Copyright Act. As a result, the judge concluded that the use of ANI’s content for AI training, as examined in this case, does not amount to copyright infringement under Indian law.
Why This AI Copyright Ruling Matters
The judgment addresses one of the most important legal questions surrounding generative AI: Can AI companies use copyrighted material to train large language models without obtaining licences?
While the ruling applies specifically to this case, it represents India’s first substantive judicial interpretation of how copyright law applies to AI model training and is likely to influence future disputes in this area.
The decision also comes as OpenAI faces similar copyright lawsuits in the United States and Canada, where courts are examining the legality of using copyrighted works for AI training. Although legal standards differ across jurisdictions, the Delhi High Court’s ruling provides an important reference point for India’s evolving AI and copyright landscape.
Key Takeaways
- The Delhi High Court ruled that OpenAI did not infringe ANI’s copyright by using its content to train ChatGPT.
- ANI failed to prove that ChatGPT memorised or reproduced its copyrighted news reports.
- The court held that AI model training is protected under the fair dealing exemption for research under India’s Copyright Act.
- This is India’s first substantive court ruling on the use of copyrighted news content for AI training.
- Similar copyright cases involving OpenAI continue in the United States and Canada.