OpenAI has accused The New York Times of allegedly hiring someone to manipulate OpenAI’s AI model, ChatGPT, to create a false premise for a copyright lawsuit. In a court filing, OpenAI claimed that The Times used deceptive tactics to generate anomalies in ChatGPT’s outputs, which were then used to support allegations of copyright infringement and audience theft against OpenAI.
OpenAI argued that The Times’ alleged actions were an attempt to mislead the court, asserting that ChatGPT is not intended to replicate Times articles and cannot be utilized in that manner by users in real-world scenarios. OpenAI also highlighted that The Times had been aware of its use of Times’ articles to train AI models since 2020 but only raised concerns when ChatGPT gained popularity in 2022.
The legal battle between OpenAI and The Times revolves around claims of copyright infringement, contributory infringement, Digital Millennium Copyright Act violations, and misappropriation. OpenAI is seeking the dismissal of these claims, citing legal infirmities, timing issues, and misinterpretations of fair use and federal laws. If OpenAI’s motion is granted, only claims of vicarious copyright infringement and trademark dilution would likely remain.
On the other hand, The New York Times contends that OpenAI unlawfully copied millions of its works without permission to develop its AI products on a significant scale. Ian Crosby, lead counsel for The Times, refuted OpenAI’s characterization of the events, emphasizing that using copyrighted material without authorization is a violation of copyright law, regardless of the purpose.
OpenAI detailed how The Times allegedly exploited bugs in ChatGPT, such as training data regurgitation and model hallucination, to create the illusion of copyright infringement. OpenAI plans to address these issues and improve its AI models but is relying on legal arguments to defend its practices.
HEALTH
WORLD AFFAIRS
TECHNOLOGY
CREATIVE
