Apple’s trade secret lawsuit against OpenAI just got a lot messier. What started in July as a straightforward accusation that two former Apple employees carried proprietary hardware knowledge to their new employer has turned into a courtroom fight over deleted messages, a leaked circuit schematic, and dueling claims about who really failed to protect confidential information. The Apple OpenAI trade dispute now sits at the center of one of the most closely watched tech legal battles of the year, and neither side is backing down.
Key takeaways
- Apple sued OpenAI and its hardware unit io Products on July 10, 2026, in the US District Court for the Northern District of California, alleging misappropriation of confidential hardware designs.
- The complaint names former Apple employees Tang Yew Tan and Chang Liu, both now working at OpenAI.
- New filings claim Liu used a confidential Apple circuit schematic and an internally named engineering tool at OpenAI, and allegedly enlisted colleague Yu-Ting Peng to destroy evidence in June.
- OpenAI, which acquired io Products in 2025 for $6.5 billion, argues Apple’s own offboarding and security failures — not theft — explain any residual file access.
- A hearing on Apple’s request for a preliminary injunction and OpenAI’s motion to dismiss is scheduled for October 1, 2026, before Judge Edward J. Davila.
Apple’s Trade Secret Case Against OpenAI Widens
Apple’s original complaint, filed July 10, 2026, accused OpenAI and its subsidiary io Products of orchestrating a coordinated effort to absorb sensitive hardware designs, manufacturing methods, and confidential engineering documents. The case landed in the US District Court for the Northern District of California, San Jose Division, and names two former Apple employees, Tang Yew Tan and Chang Liu, who both left Apple to join OpenAI. Apple frames their departure as part of a deliberate strategy by OpenAI to pull proprietary knowledge out of one of its biggest rivals in consumer hardware.
New Evidence From Chang Liu’s Apple Laptop
The dispute escalated in late August when Liu’s legal counsel handed over his old Apple work laptop for investigation. Apple says the device shows Liu used a confidential Apple circuit schematic in his work at OpenAI, along with a tool that shares a name with an internal Apple engineering application. According to a court filing reported by TechCrunch, Apple alleges OpenAI was “well-aware” of Liu’s continued access to Apple data and that Liu asked an OpenAI colleague, Yu-Ting Peng, to help destroy evidence in June once he learned Apple was investigating him.
Apple’s filing put it bluntly: “The MacBook represents the very limited information Defendants provided so far (and only after weeks of delay), and shows Apple is not conducting ‘fishing expeditions’ but that its trade secrets are being used and evidence is being destroyed.” Earlier filings in the case had already included text messages from Liu, punctuated with “crying laughing” emojis, that Apple says prove he knew he still had access to company files after leaving.
OpenAI Pushes Back and Seeks Dismissal
OpenAI’s counter-argument does not dispute the facts head-on — it reframes them entirely. The company says any lingering access to Apple’s systems reflects a security lapse on Apple’s side, not a theft scheme engineered by OpenAI. That framing has become the backbone of OpenAI’s defense across multiple filings.
The “Residual Access” Defense
In an August 4, 2026 blog post, OpenAI wrote: “Apple now tries to shift the blame to ‘residual access,’ but they also don’t disclose that this is a common issue with Apple which is caused by them failing to properly manage system access when people leave.” OpenAI’s position is that Liu only accessed Apple files after leaving the company because former colleagues asked for his help, not to funnel secrets into hardware development. Apple has rejected that explanation, claiming Liu instead “exploited a rare, previously unknown authentication bug” to keep downloading confidential files after his departure.
OpenAI has also gone on offense procedurally. In a reply filed August 26, 2026, in support of its motion to dismiss, attorneys for OpenAI Foundation, OpenAI Group PBC, io Products LLC and Tang Yew Tan argued that Apple’s complaint still fails to identify any actual trade secret, fails to plead misappropriation by any named defendant, and fails to show real harm. The filing, submitted in case number 5:26-CV-07078-EJD before Judge Edward J. Davila, contends Apple relies on “vague categories and catch-alls” rather than pinpointing specific stolen information — a legal distinction that could determine whether the case survives past the pleading stage. As of September 1, 2026, OpenAI maintained that Apple still had not demonstrated that any confidential information was actually stolen, only that suspicious circumstances existed.
A $6.5 Billion Hardware Bet Behind the Dispute
The timing of OpenAI’s hardware ambitions is central to why this case matters beyond a personnel dispute. OpenAI’s io Products acquisition, completed in 2025 for $6.5 billion, signaled the company’s intent to build its own consumer devices rather than remain a software layer riding on other companies’ hardware. According to Apple’s initial filing, more than 400 former Apple employees now work at OpenAI — a talent migration that Apple’s lawsuit suggests is not coincidental, though OpenAI disputes that characterization.
What makes the standoff unusual is that Apple and OpenAI are still commercial partners. ChatGPT remains integrated into Apple products, an arrangement both companies have publicly promoted even as their lawyers trade accusations in federal court. That contradiction is part of what makes this Apple OpenAI trade dispute so closely watched across the tech industry: two companies can compete fiercely over engineering talent and hardware ambitions while continuing to build products together.
Why this matters: the case will likely test whether knowledge that employees carry with them counts as a protectable trade secret or simply professional experience that naturally travels between employers. That distinction has implications far beyond Apple and OpenAI, touching any company worried about talent poaching in a hot AI hardware market.
What’s Next: Injunction Hearing on October 1
Apple has asked the court for a preliminary injunction that would block OpenAI from continuing hardware work built on Apple’s technology while the case proceeds, along with expedited discovery to identify whether more former employees are implicated. OpenAI, in turn, wants the entire complaint dismissed before it gets that far.
Both motions converge on the same date. The hearing is scheduled for October 1, 2026, at 9:00 a.m. Pacific time in Courtroom 4, with Judge Edward J. Davila presiding. That single hearing will show how seriously the court weighs Apple’s request to freeze OpenAI’s hardware activities — and whether Apple’s trade secret claims are specific enough to survive OpenAI’s dismissal push. No trial date has been set.
FAQ
What are the core allegations Apple makes against OpenAI?
Apple alleges that OpenAI and its subsidiary io Products orchestrated a coordinated effort to misappropriate sensitive hardware designs and confidential engineering documents, aided by former Apple employees now working at OpenAI.
How does OpenAI respond to Apple’s trade secret accusations?
OpenAI argues that Apple’s own security and offboarding failures allowed former employees to retain access to confidential files, making it a security issue rather than theft. OpenAI also states that Apple has not provided sufficient evidence of actual theft.
What is the significance of OpenAI’s acquisition of io Products in this dispute?
The $6.5 billion acquisition of io Products in 2025 signaled OpenAI’s serious intent to compete in consumer hardware, which Apple claims motivated OpenAI’s alleged strategy to acquire proprietary hardware knowledge.
What is the current status of the legal proceedings between Apple and OpenAI?
Apple has requested a preliminary injunction to halt OpenAI’s hardware work, with a hearing scheduled for October 1, 2026, while OpenAI has pushed for dismissal of the claims citing insufficient evidence.
Article produced with the assistance of artificial intelligence and reviewed by the editorial team.
Be the first to comment