Alec Radford, a key researcher behind many of OpenAI’s groundbreaking AI technologies, has been subpoenaed in a copyright lawsuit agains the AI startup, according to a court filing on Tuesday.
The filing, submitted by an attorney for the plaintiffs in the U.S. District Court for the Northern District of California, states that Radford was served with a subpoena on February 25.
Radford, who departed OpenAI late last year to focus on independent research, played a crucial role in the development of generative pre-trained transformers (GPTs), the technology powering OpenAI’s most well-known products, including ChatGPT. He was the lead author of OpenAI’s foundational research paper on GPTs.
Having joined OpenAI in 2016, Radford contributed to multiple GPT models as well as other projects like Whisper, the company’s speech recognition model, and DALL-E, its AI-powered image generator.
The copyright lawsuit, titled re OpenAI ChatGPT Litigation, was filed by book authors such as Paul Tremblay, Sarah Silverman, and Michael Chabon. The plaintiffs allege that OpenAI infringed their copyrights by using their written works to train its AI models. Additionally, they argue that ChatGPT reproduces sections of their works without proper attribution.
In a ruling last year, the court dismissed two of the plaintiffs’ claims against OpenAI but allowed the case for direct copyright infringement to proceed. OpenAI, however, asserts that its use of copyrighted data for AI training falls under the fair use doctrine .
Radford is not the only prominent figure targeted in this legal battle. Attorneys for the plaintiffs are also seeking to depose Dario Amodei and Benjamin Mann, both former OpenAI employees who later co-founded the AI startup Anthropic. Amodei and Mann have opposed the motions, arguing that the demands are excessively burdensome.