Second Mathematician Accuses OpenAI of Dishonesty Over Math Training Data

A second mathematician, Andreas Thom, has publicly accused OpenAI of unethical and dishonest behavior regarding the origins of its AI training data. Thom raised his concerns in a series of Mastodon posts, suggesting that pre-announcement interactions he and colleagues had with ChatGPT may have contributed to OpenAI's recent mathematical breakthroughs. His allegations follow a separate dispute that erupted just days earlier over whether OpenAI's models benefited from unpublished mathematical research. Researchers are now calling on the company to provide greater transparency about the data used to train its models. OpenAI has yet to publicly address the specific claims made by Thom and others in the mathematics community.