Photo: Koshiro K/Shutterstock.com
OpenAI, the US-based artificial intelligence leader and developer for the ChatGPT language model, has requested the Delhi High Court to dismiss allegations of copyright infringement brought forward by the Federal of Indian Publishers (FIP), a group representing book publishers. The company contends that its AI services only rely on publicly available information.
FIP represents prominent Indian publishers alongside international names such as Penguin Random House and Bloomsbury and accuses OpenAI’s ChatGPT of generating summaries and extracts from unauthorised digital copies of books. This, they argue, harms their business interests by bypassing legitimate licensing agreements, reports Reuters.
Pranav Gupta, secretary of the Federation, expressed concerns to Reuters, stating that much of the book-related content generated by ChatGPT is derived from websites with normal licensing arrangements with publishers. However, OpenAI refutes these allegations, emphasising in its court filing dated January 26 that its web-crawling technologies only access publicly available data.
The company argued that publishers have failed to demonstrate even a single instance of its AI models being trained on original copyrighted works.
In its 21-page response, OpenAI explained that its training sources include platforms such as Wikipedia, as well as publicly available abstracts, summaries, and tables of content published online. The company assets that this practice is protected under fair use principles, a key legal argument it has used in similar cases globally.
“Web crawlers are designed to only access publically available data,” OpenAI said. The book publishers have “entirely failed to demonstrate even a single instance” that OpenAI trained on “original literary work.”
Adding another layer of complexity to the case, OpenAI has also argued that Indian courts lack jurisdiction over its operations since its servers are outside India. This defence mirrors strategies adopted by other tech companies facing international legal challenges over AI practices.
This case is just one of many being heard globally, as content creators, including authors, news organisations, and musicians, have accused tech companies of using their copyrighted work to train AI systems without proper licensing. Nearly a dozen Indian digital media outlets, including those owned by tycoons Gautam Adani and Mukesh Ambani, have joined the case against OpenAI.
In the News: Apple patches actively exploited zero-day bug
