Online news will disappear, only ChatGPT will remain: Media to Delhi High Court in OpenAI case

The Digital News Publishers Association (DNPA) on Tuesday argued before the Delhi High Court that OpenAI was infringing the rights of media organizations by training its Large Language Model (LLM) ChatGPT on the basis of online news reports.

Senior Advocate Rajshekhar Rao, appearing for the DNPA, said that circulation of physical newspapers has already come down drastically and ChatGPT would make even online news vanish.

"Physical newspapers circulation has come down drastically. Physical newspapers are disappearing, digital news will disappear and only ChatGPT will remain. It reduces my ability or incentive to publish," he said.

The submission was made before Justice Amit Bansal during the hearing of Asian News International (ANI)'s copyright infringement suit against OpenAI.

DNPA is an intervenor in the case and it represents India's leading media organizations including the Times Group, The Hindu, The Indian Express, NDTV and others.

"If I have the exclusive right to authorize you to do this or not to do this, the fact that you pick it up from the public domain, the fact that you use it to train ... the minute you download without my authority, you are infringing my exclusive right," Rao submitted.

Rao argued that storage of the news reports even for a "transient moment" will amount to infringement of copyright of the news organizations.

"Storage of an infringing copy itself is infringing. Even if you store for a transient moment you are infringing. You cannot use my copyright without my license," he said.

Highlighting a particular instance of alleged copyright violation, the senior counsel submitted that ChatGPT had provided excerpts of an interview on a query put by his colleague.

"Any reproduction which is not consistent with what the Act authorizes or permits is an infringing copy," he added.

Rao also argued that even for research, OpenAI needs to first obtain a lawfully acquired copy.

"So you have to buy SCC online to be able to crack it in the privacy of your home for research purposes," he said.

At this, the Court referred to the amendments in Copyright Act, 1957.

"So this a classic case where a law which was last amended in 2012 is trying to keep pace with technology," Justice Bansal remarked.

Rao will continue his arguments on next date of hearing. The next three dates of hearing in the case are August 18, September 12 and September 23.

ANI filed the suit alleging that its original content was being exploited by OpenAI for commercial gain and to train ChatGPT for answering the user queries. In response, OpenAI has said that copyright protection in the news reporting is very narrow as there exists a greater public interest in dissemination of information.