OpenAI has requested a federal judge to dismiss certain claims within the New York Times' copyright lawsuit, alleging that the newspaper manipulated its ChatGPT chatbot and other AI systems for the creation of distorted evidence. In a recent legal filing, OpenAI accuses the New York Times of using "deceptive prompts" that contravene the AI firm's terms of use, thereby misleadingly causing the reproduction of its content.

The company criticized the lawsuit for not adhering to the high journalistic standards the Times is known for, stating, "The Times paid someone to hack OpenAI's products." However, OpenAI stopped short of accusing the newspaper of violating any anti-hacking legislation and did not identify the individual allegedly hired for this purpose.

In response, Ian Crosby, attorney for the New York Times, rejected the notion that their actions constituted hacking, asserting that they were merely examining whether OpenAI's products unlawfully replicated the newspaper's copyrighted material.

The lawsuit, filed against OpenAI and its main investor Microsoft in December, charges them with unauthorized use of the newspaper's articles to train their chatbots. This case is part of a broader legal battle involving several copyright holders suing tech firms over AI training's alleged misuse of copyrighted content.

The tech industry argues that their AI models utilize copyrighted materials under fair use provisions, warning that these legal actions could hinder the sector's development. Yet, the definitive legal stance on whether AI training constitutes fair use remains undecided, with courts so far dismissing claims of infringement due to insufficient proof that AI-generated outputs mimic copyrighted materials directly.

In its legal defense, OpenAI contends that the anomalies pointed out by the New York Times were the result of extensive manipulation and do not reflect typical ChatGPT usage. The company also maintains that AI models should be free to learn factual information, similar to how news organizations commonly share and report on the same stories.