Apple Gets Early Look at OpenAI Hiring in Secrets Case

U.S. District Judge Edward Davila has granted Apple early access to OpenAI's hiring materials in its trade secrets lawsuit. The order, reported Oct. 13, opens a window into how OpenAI recruited and interviewed former Apple staff for its hardware work. TechCrunch

Apple filed the case in July 2026 against OpenAI and two former employees. The complaint accuses OpenAI of running a broad effort to systematically acquire and exploit Apple's confidential information. Reuters

The early evidence, known legally as expedited discovery, covers interviews with former Apple employees who applied for jobs in OpenAI's consumer hardware and robotics groups. It includes checklists, guides, instructions, manuals, scripts, technical presentations, and any records showing interview discussions that referred to Apple technology or components.

OpenAI has five days to hand over materials it has already collected that match that request, and 14 days for other covered materials.

Davila also granted Apple access to private LINE messages involving people including OpenAI employees Chang Liu and Alyssa Peng. LINE is a consumer messaging app outside corporate email systems. He approved depositions, which are sworn interviews taken before trial, for Liu, Peng, and OpenAI chief hardware officer Tang Yew Tan. All three previously worked at Apple. The earlier complaint named Chang Liu and Tang Tan as the two individual defendants. 9to5Mac

Apple's core claims concern taken files and recruiting. The company says former employees used Apple laptops to download confidential material on unannounced technologies, features, and products, including technical specifications, engineering presentations, and proprietary project data. It also says former Apple employees now at OpenAI coached others who were considering leaving Apple about what information to bring to interviews.

The judge did not grant everything Apple requested early. He denied early access to OpenAI's research and development on metal finishes and power management. He also denied broader early requests for OpenAI hardware research and for information about what its former employees created while at OpenAI. He left open the possibility of narrower versions of those requests in normal discovery.

Both sides have already spoken publicly. OpenAI published a response on Aug. 3 titled "Apple is getting this wrong." Apple sought a preliminary injunction, a court order to block certain actions before trial, against OpenAI and two former employees on Aug. 4. Reuters

The broader context here is how trade secret cases often turn on hiring when a software company builds a hardware team. Courts often allow early access to recruiting records and interview questions because that evidence helps show whether confidential information moved with people or was developed independently. Scripts, checklists, and messages between recruiters and hiring managers matter for that question.

In my view, Davila's split decision follows a pattern I have seen since the PC era in technology disputes over alleged misuse. Allowing interview discovery lets Apple try to establish a factual basis for misuse. Holding back research discovery until that basis exists limits early looks into a competitor's product plans. For hardware and AI teams, it is worth noting in plain terms that the inclusion of LINE messages is a reminder that personal messaging apps do not stay outside court orders once specific people are named.

What to watch next is what the interview records actually show. A mention of Apple technology in an interview can mean different things, from a candidate describing past work as background to an interviewer asking for specifics. The depositions Davila has now allowed will be needed to separate those cases. If Apple can link specific disclosures to later OpenAI design choices, its case for broader hardware evidence gets stronger. If not, OpenAI will have limited early disclosure to its hiring process.

Over the longer arc, either result still leaves room to keep building. Clear rules for interviews and careful onboarding for new hires, where new teams avoid using a prior employer's confidential material, protect both departing engineers and hiring companies. That approach reduces the need for courts to review product work document by document.