Lawyers from The New York Times and Daily News are suing OpenAI for allegedly using their works to train AI models without permission. Those lawyers claim OpenAI engineers accidentally erased relevant data needed for the lawsuit. OpenAI agreed to provide virtual machines so that the publishers’ lawyers could search for their copyrighted content in OpenAI’s AI training sets. The lawyers and their hired experts spent over 150 hours since November 1, 2024 searching OpenAI’s training data.
However, OpenAI engineers accidentally erased all of the publishers’ search data, resulting in the loss of an entire week’s worth of their lawyers’ and experts’ work. OpenAI tried to recover the data, but the recovered data cannot be used to locate the news plaintiffs’ copied articles. OpenAI insisted that the incident was unintentional, but the plaintiffs’ lawyers believe OpenAI should search its datasets using its own tools. OpenAI’s lawyers denied deleting any evidence and placed blame on the plaintiffs’ team for a system misconfiguration. OpenAI believes that training models using publicly available data is fair use. It has also secured licensing deals with a number of new publishers. OpenAI has not confirmed or denied training its AI systems on copyrighted works without permission.
The ainewsarticles.com article you just read is a brief synopsis; the original article can be found here: Read the Full Article…
