Indian news agency ANI is suing ChatGPT for using its content without permission

OpenAI is fighting several lawsuits in the US from publications for unauthorized use of their content in its models, and it looks like these lawsuits have reached India as well.

Indian news agency ANI has filed a lawsuit in a Delhi court against ChatGPT maker OpenAI for using its content without permission. ANI alleges that OpenAI used ANI’s published content to train its models without its express permission. ANI also accused OpenAI of attributing fake news to the agency.

ANI says it typically licenses its reporting to news organizations such as The Financial Times and The Associated Press for a fee. But it claims that OpenAI used its content for free to train its ChatGPT models. ANI says this is infringing its copyright and damaging its business interests. ANI also alleged that some of the content created by ChatGPT included false claims that the articles were published by ANI, which undermined its credibility. ANI is seeking an injunction to stop OpenAI from using their material, and to address the alleged damages caused by the improper distribution of the fabricated content.

OpenAI, for its part, denies the allegations. An OpenAI representative said the company builds its AI models using publicly available data and uses fair use principles. OpenAI also emphasized that it has stopped using ANI content for future ChatGPT training. According to OpenAI, ANI content has been part of an internal block list since September, effectively preventing it from being used in any future AI model training.

But the fact that OpenAI blacklisted ANI’s content in September suggests that its content may have been in the company’s possession by then. This can be a difficult situation for many companies that use publicly available data to train their models. In the US, the New York Times and the Chicago Tribune similarly sued OpenAI, claiming that their copyrights were infringed when these AI models were trained on their articles.

However, OpenAI and other AI companies say they’ve used large parts of the public web to train their models, but the models typically don’t repeat this back to their users verbatim — instead, they use the resulting data to learn how the world works and then use this information in your answers. They say it is similar to how a person reads many articles and books, but then writes an original essay based on his knowledge.

It will be up to the courts to decide how original ChatGPT’s responses are. But the Delhi court — and other courts around the world — will have to decide whether AI companies training their models on public data is infringing copyright. This is a new legal issue brought about by the development of new technologies, and the outcome of how the courts deal with this situation could have major implications for the field of AI as a whole.