New York Times alleges OpenAI deleted potentially incriminating evidence

Background

The lawsuit, filed against OpenAI and Microsoft last year, alleges that the companies had illegally used The New York Times’ articles to train artificial intelligence tools like ChatGPT. As part of the discovery process, OpenAI was required to show The New York Times its training data, which included a “sandbox” of two virtual machines that the paper’s lawyers could sift through.

The Incident

Reactions

OpenAI spokesperson Jason Deutrom stated that the company disagrees with the characterizations made and will file its own response soon. The New York Times declined to comment.

Context

The incident has sparked concerns about the reliability of AI companies’ data and the potential for evidence to be deleted or tampered with. It also raises questions about the accountability of these companies in their use of copyrighted material.

Key Points

• The New York Times alleges that OpenAI’s engineers inadvertently erased data the paper’s team spent more than 150 hours extracting as potential evidence.

• OpenAI was able to recover much of the data, but the original file names and folder structure were lost.

• The missing information cannot be used to determine where the news plaintiffs’ copied articles may have been incorporated into OpenAI’s artificial intelligence models.