In a federal courtroom in California on Thursday, Elon Musk testified that his AI startup, xAI, used OpenAI’s models to improve its own Grok technology. The admission came during legal proceedings addressing model distillation, an industry-standard practice where a larger AI model serves as a "teacher" to transfer knowledge to a smaller "student" model.

While model distillation is frequently employed internally by companies to train new models, smaller AI labs also leverage it to replicate the performance of larger competitors' models. Musk’s testimony did not specify which OpenAI models were utilized or the extent of their contribution to Grok’s development.

The courtroom exchange followed a broader legal dispute involving allegations of improper use of proprietary AI technologies. Model distillation itself is not inherently illegal but has drawn scrutiny in cases involving unauthorized access to protected models.

Musk’s statement provides new insight into xAI’s training methodology, though further details about the collaboration or technical specifics remain undisclosed.

Source: The Verge