OpenAI, led by Sam Altman, has uncovered evidence suggesting that a former employee took over 100 confidential files concerning an unreleased device when he joined Apple. This revelation came to light in a significant court filing on Wednesday as part of the ongoing trade secret dispute between the two companies.
The allegations were made after what OpenAI described as a “recent and ongoing forensic investigation” linked to the lawsuit. Apple initiated legal action against OpenAI and two past employees back in July, claiming they had stolen various trade secrets, including numerous confidential files related to its hardware.
In its filing, OpenAI claimed that the former employee had “sent over a hundred highly confidential files about at least one unreleased OpenAI device” to his personal accounts. These files reportedly contained “detailed images” as well as “highly confidential engineering and manufacturing data” pertaining to the device or devices involved.
According to the filing, some of these files were transmitted just after midnight in Singapore on June 2, 2026 — which was the former employee’s last day at OpenAI. It further stated that this individual had made a presentation to Apple’s senior leadership, specifically mentioning Apple’s vice president of operations in Asia, about OpenAI’s product plans.
The employee had a short stint at OpenAI, about six months, after previously working at Apple, before returning there. OpenAI has alleged that these documents contributed to Apple’s efforts in launching “AI-enabled products,” particularly noting the recent introduction of the “Apple Home” line of smart home devices.
OpenAI attempted to discuss the matter with Apple but reported that it hasn’t received the “requested assurances” that Apple, under CEO John Ternus, has not utilized or retained the documents in question.
The filing expressed concerns over the potential misuse of OpenAI’s trade secrets within Apple, warning that Apple employees could use OpenAI’s confidential innovations to influence upcoming products.
When reached for a statement, an Apple spokesperson reiterated the company’s strict policies designed to prevent new hires from misusing confidential information from former employers. They also mentioned that Apple did not find any evidence supporting OpenAI’s claims and had not responded further to requests for clarification.
This allegation emerged on the same day as a pivotal hearing in the case. U.S. District Judge Edward Davila signaled a likelihood of siding with Apple on its request for an injunction to prevent ex-Apple employees now at OpenAI from utilizing proprietary information, according to reports.
However, Judge Davila also seemed to downplay Apple’s proposal for an extensive forensic examination of all OpenAI devices and code repositories, categorizing it as a “huge ask.” In response, Apple stated it was willing to confine the investigation to a select few former employees who transitioned to OpenAI.
Apple’s lawsuit indicated that two individuals, Chang Liu, a previous senior system electrical engineer, and Tang Yew Tan, a former VP of product design, were implicated in allegedly taking company intellectual property when they left.
It was claimed that Liu did not return his Apple-issued laptop and employed a software bug to access confidential files after his departure. The lawsuit also mentioned Liu’s seemingly cavalier attitude towards that access, as he reportedly left a message on a colleague’s work laptop expressing amusement over it.
Tan was accused of “methodically using Apple’s confidential information to benefit OpenAI,” which included emailing himself data about the company’s suppliers before leaving Apple.
OpenAI has categorically denied these allegations, insisting that it has “no interest in other companies’ trade secrets.”
