OpenAI has requested a judge to dismiss parts of The New York Times’ lawsuit after accusing the publisher of using unethical methods to get evidence.
The tech firm alleges that the Times “paid someone to hack” into its products, including ChatGPT, to create 100 instances of copyright infringement.
The AI startup submitted the legal documents to a Manhattan federal court.
It claims the Times achieved these results through “tens of thousands of attempts” using “deceptive prompts”
This, it’s claimed, directly contradicts OpenAI’s usage terms.
OpenAI argues the tactics used by the Times, could be categorized as prompt engineering or “red-teaming.”
It says these are not standard practices among regular users.
Despite being a routine procedure among AI safety teams and researchers to identify AI vulnerabilities, OpenAI suggests the Times’ approach was specifically aimed at violating its terms of service.c
“The scale of OpenAI’s copying is much larger than the 100-plus examples set forth in the complaint”
Ian Crosby, lead counsel for the Times, said: “In this filing, OpenAI doesn’t dispute — nor can they — that they copied millions of The Times’s works to build and power its commercial products without our permission.
“What OpenAI bizarrely mischaracterizes as ‘hacking’ is simply using OpenAI’s products to look for evidence that they stole and reproduced The Times’s copyrighted works.
“And that is exactly what we found. In fact, the scale of OpenAI’s copying is much larger than the 100-plus examples set forth in the complaint.”
This legal confrontation is part of a larger dispute involving using copyrighted material for AI training data.
The Times filed the lawsuit in December and is seeking significant damages from OpenAI and Microsoft.
OpenAI previously said modern AI models rely heavily on copyrighted works for training.
Looking to boost your online brand? Create your FREE business profile at WhatBiz? Here
OpenAI wrote in a filing last month: “Because copyright today covers virtually every sort of human expression—including blog posts, photographs, forum posts, scraps of software code, and government documents—it would be impossible to train today’s leading AI models without using copyrighted materials.”
OpenAI CEO Sam Altman said he was “surprised” over the Times’ lawsuit and claimed the company does not specifically need the Times’s data.
He said: “We actually don’t need to train on their data. I think this is something that people don’t understand.
“Any one particular training source, it doesn’t move the needle for us that much.”
The company has initiated dialogues with various publishers to secure content licensing agreements and has partnered with several media houses.
It said: “We expect our ongoing negotiations with others to yield additional partnerships soon.”
