OpenAI vs ANI: IP, copyright laws, 'fair use' doctrine likely to be tested

OpenAI vs ANI in Delhi HC on Friday: Experts believe decision will define working, future of AI firms in India

The proceedings as well as the court decision in the dispute between ANI Media and ChatGPT maker OpenAI could define the future of AI companies in India, according to experts on legal matters, artificial intelligence, and data privacy.

  The Delhi High Court will continue hearing on Friday the case moved by ANI Media against OpenAI, in which the former has alleged that its content had been illegally used to train large language models (LLM) and other artificial intelligence (AI) models.

  Apart from ANI, several other domestic news publishers and their representative associations, media houses, and music labels have sought to intervene in the case as petitioners.

  The case, according to experts, sits at the crucial juncture of intellectual-property (IP) laws, the evolution of AI, copyright laws, IP infringement, and the boundaries of “fair use” defence available to firms. “The challenge is to maintain a balance between protecting intellectual properties and fostering AI-driven innovation. A rigid, over-restrictive approach could stifle technological advancement, while an unchecked AI replication of copyrighted material could erode creators’ rights,” said Jameela Sahiba, associate director of AI and public affairs at technology policy advocacy group The Dialogue. 

  ANI’s plea is the latest in the ongoing worldwide tussle between, on one side, news publishers, content creators, music publishers and labels, and, on the other hand, AI tools and LLM developers such as OpenAI and Ross Intelligence, an AI startup. 

  In the United States (US), The New York Times had filed a case against OpenAI for allegedly using its content illegally to train the LLMs that power ChatGPT and other tools, while Reuters won a case in the US after a federal judge ruled that Ross Intelligence used the news agency’s copyrighted material illegally and unlawfully. 

Courts in India will decide the case on three major parameters, said Swati Sharma, a partner, and head of intellectual property practice at law firm Cyril Amarchand Mangaldas. 

“The primary issue to be decided by the court is whether the storage and use of copyrighted data and the generation of responses thereof constitute infringement of copyright,” Sharma said.

  The other two crucial aspects, she said, would be deciding whether OpenAI’s use of copyrighted data qualifies as “fair use” under Section 52 of the Copyright Act, 1957, and whether Indian courts have jurisdiction over the matter since the servers of the defendant are located in the US.

  Salman Waris, founder of law firm TechLegis, said the important aspect of fair law, however, would be determined differently in India compared to the US.   

THE CONTROVERSY

- ANI alleged its content was illegally used to train large language models (LLM) and other AI models

- Several domestic news publishers and media houses have sought to intervene in the case as petitioners

- Earlier, The New York Times had filed a case against OpenAI for using its content illegally to train the LLMs

- News agency Reuters had won a case in the US after an AI startup used its copyrighted material illegally