OpenAI may have made a fatal misstep in copyright fight with news orgs

Key Points:

• News organizations led by The New York Times accuse OpenAI of deliberately concealing evidence by misleading the court about its ability to search ChatGPT logs, which are crucial to proving whether OpenAI infringed copyrights or used content under fair use.
• OpenAI allegedly hid large anonymized datasets of ChatGPT logs—totaling around 80 million entries—that it had already searched internally, while claiming it was too burdensome to provide them to plaintiffs, thereby prolonging discovery and increasing costs.
• The court found the smaller 20 million log sample provided by OpenAI heavily redacted and effectively unusable, while OpenAI reportedly deleted or compressed billions of other logs, violating preservation orders and obstructing evidence access.
• News plaintiffs are seeking serious sanctions against OpenAI, including barring the company from using the disputed 20 million log sample and instructing the jury that OpenAI deleted significant evidence, arguing these measures are necessary due to OpenAI’s intentional misconduct.
• OpenAI denies wrongdoing, framing the sanctions motion as an attempt to invade user privacy and maintain that their use of copyrighted material falls under fair use, while the outcome of this discovery dispute could critically impact the ongoing copyright lawsuit.