OpenAI has refuted claims made by several Indian media houses that it utilised their content without authorisation to train ChatGPT. These allegations stem from a copyright lawsuit originally filed by ANI, a well-known Indian news agency. According to Reuters, OpenAI is actively seeking to prevent major Indian media companies, including those linked to prominent businessmen Gautam Adani and Mukesh Ambani, from joining the ongoing legal battle.
Media organisations such as NDTV, Network 18, The Indian Express, Hindustan Times, and the Digital News Publishers Association (DNPA) have accused OpenAI of scraping their content without consent. In January, these media outlets approached the Delhi High Court, requesting to be included in the existing lawsuit against the AI firm.
However, OpenAI, in a recent 31-page legal submission dated February 11, contends that it is not legally required to establish agreements with media entities to use publicly available data. The company has categorically denied the accusations, asserting that neither ANI’s content nor that of the DNPA members has been used to train its AI systems.
ANI had initially taken legal action in November 2024, alleging that ChatGPT reproduces its published material without permission. At that time, OpenAI defended its practices, stating that its models are developed using publicly accessible data, within the framework of fair use and established legal precedents.
Meanwhile, in December 2024, the Federation of Indian Publishers (FIP) also challenged OpenAI in court, demanding that the company halt its use of copyrighted material, including books and digital content. The FIP, which represents publishers such as Bloomsbury, Penguin Random House, Cambridge University Press, Pan Macmillan, Rupa Publications, and S. Chand and Co., insisted that OpenAI should either negotiate licensing agreements or remove the disputed datasets.
Copyright lawsuits against AI firms are not new. In 2023, The New York Times sued OpenAI and its partner Microsoft, alleging unauthorised use of its content for training large language models (LLMs). Later that year, authors Nicholas Basbanes and Nicholas Gage also filed a class-action lawsuit, accusing the companies of unlawfully using their copyrighted works. Additionally, Meta faced similar legal action from a group of authors, who claimed that its LLaMA AI models were trained on pirated content with the knowledge of CEO Mark Zuckerberg.
