“It is not lost on the Court the important role of datasets in AI training and development, and that, hypothetically, datasets and details about them could be trade secrets,” Bernal wrote. But xAI “has not alleged that it actually uses datasets that are unique, that it has meaningfully larger or smaller datasets than competitors, or that it cleans its datasets in unique ways.”
Therefore, xAI is not likely to succeed on the merits of its Fifth Amendment claim.
The same goes for First Amendment arguments. xAI failed to show that the law improperly “forces developers to publicly disclose their data sources in an attempt to identify what California deems to be ‘data riddled with implicit and explicit biases,’” Bernal wrote.
To xAI, it seemed like the state was trying to use the law to influence the outputs of its chatbot Grok, the company argued, which should be protected commercial speech.
Over the past year, Grok has increasingly drawn global public scrutiny for its antisemitic rants and for generating nonconsensual intimate imagery (NCII) and child sexual abuse materials (CSAM). But despite these scandals, which prompted a California probe, Bernal contradicted xAI, saying California did not appear to be trying to regulate controversial or biased outputs, as xAI feared.
“Nothing in the language of the statute suggests that California is attempting to influence Plaintiff’s models’ outputs by requiring dataset disclosure,” Bernal wrote.
Addressing xAI’s other speech concerns, he noted that “the statute does not functionally ask Plaintiff to share its opinions on the role of certain datasets in AI model development or make ideological statements about the utility of various datasets or cleaning methods.”
“No part of the statute indicates any plan to regulate or censor models based on the datasets with which they are developed and trained,” Bernal wrote.
Public “cannot possibly” care about AI training data
Perhaps most frustrating for xAI as it continues to fight to block the law, Bernal also disputed that the public had no interest in the training data disclosures.