OpenAI has claimed that The New York Times «paid someone to hack into OpenAI products such as ChatGPT to set up a lawsuit» against the leading AI vendor. In the lawsuit filed, OpenAI claims that «100 examples in which a particular version of OpenAI’s GPT-4 model allegedly generated several paragraphs of Times content in response to user queries» do not reflect how ordinary people use ChatGPT.
Instead, it allegedly took the Times «tens of thousands of attempts to generate» these alleged «highly anomalous results» by «targeting and exploiting a bug» that OpenAI now «commits to fixing».
Contrary to the allegations in the complaint, ChatGPT is in no way a substitute for a subscription to The New York Times. In the real world, people do not use ChatGPT or any other OpenAI product for this purpose. Nor could they. Certainly, no one can use ChatGPT to receive Times articles at will.
— OpenAI argues in its motion, which seeks to dismiss most of The Times’ claims.
OpenAI argues that the court should dismiss the claims for direct copyright infringement, contributory copyright infringement, violation of the Digital Millennium Copyright Act, and misappropriation, all of which it calls «legally untenable».
If OpenAI’s motion is granted, then most likely only the trademark dilution claims will remain. But if The Times wins, which is entirely possible — OpenAI may be forced to remove ChatGPT and start over, transmits ArsTechnica..