Articles

Apple Sues OpenAI: Trade Secret Lawsuit Explained

Apple sued OpenAI, io Products and two ex-employees for trade secret theft over AI hardware. Here are the allegations, the players, and what's at stake.

Chisato Chisato · · 6 min read
An abstract swirl of generative light in blue and violet

The two companies that once shared a stage now share a courtroom. On July 10, 2026, Apple filed suit in federal court in the Northern District of California, accusing OpenAI, the Jony Ive-led hardware startup io Products, and two former Apple employees of stealing trade secrets to jump-start OpenAI’s push into consumer hardware. The complaint is unusually blunt: Apple alleges that OpenAI’s device business is, in its words, “rotten to its core by its illegal reliance on misappropriated trade secrets.”

What Apple is alleging

Apple’s central claim is that the theft was not the work of a rogue employee but a coordinated effort. “At every level, from members of its Technical Staff to its Chief Hardware Officer, and in coordination with business partners, OpenAI has been stealing Apple’s trade secrets and confidential information,” the complaint states. Apple frames the conduct as “a coordinated pattern of misconduct at an institutional level” and brings claims for trade secret misappropriation and breach of contract.

The suit names four defendants: OpenAI, io Products, and two former Apple engineers who left for OpenAI — Tang Yew Tan and Chang Liu. Tan, a former Apple vice president, is now OpenAI’s Chief Hardware Officer, which is why Apple’s “at every level” language reaches all the way to the top of OpenAI’s hardware organization.

The two named employees

Apple’s most specific allegations center on the two individuals.

Tang Yew Tan is accused of turning OpenAI’s hiring process into an intelligence-gathering operation. According to the complaint, Tan “directed job candidates still working for Apple to bring ‘actual parts’ from Apple to their interviews for ‘show and tell’ sessions” — sessions in which he and his OpenAI team could, Apple says, elicit still more confidential information from people who were, at the time, Apple employees. Apple further alleges that OpenAI circulated “a checklist that Tang put together” to help new hires evade detection from Apple’s security teams as they departed.

Chang Liu, described as a former senior systems electrical engineer at Apple, is accused of more conventional data theft. Apple claims Liu took an Apple laptop and, while developing hardware for OpenAI, accessed and downloaded dozens of confidential files from Apple’s network before leaving. Where the allegations against Tan describe a system for extracting knowledge, the allegations against Liu describe the removal of specific documents and property.

io Products and the hardware ambitions in question

The suit is inseparable from OpenAI’s push into physical devices. Last year, OpenAI acquired io Products, the hardware startup founded by former Apple design chief Jony Ive, in a deal valued at roughly $6.4 billion. That acquisition signaled OpenAI’s intent to build its own consumer AI hardware — a category Apple has dominated for two decades — and it is precisely the effort Apple now says was built on stolen foundations. The complaint names io Products as complicit in the alleged scheme rather than an innocent bystander.

The context makes the filing especially charged. Apple and OpenAI were partners in the recent past: OpenAI’s models were woven into Apple’s software, and the two firms were, for a time, the marquee example of a device maker teaming with a frontier lab. Relations cooled once OpenAI moved to enter the hardware business directly. Apple’s decision to license Google’s models for its assistant — the Siri redesign built around Gemini — underscored how far the two companies had drifted before the lawsuit made the split explicit.

OpenAI’s response

OpenAI has pushed back, if briefly. In a statement, the company said, “We have no interest in other companies’ trade secrets. We remain focused on building innovative technology that empowers people everywhere.” Drew Pusateri, OpenAI’s director of strategic communications, echoed the line publicly, saying the company has “no interest in other companies’ trade secrets.”

Apple, for its part, says it tried to raise concerns before litigating and that OpenAI “never responded.” That detail matters legally: companies pursuing trade secret claims often point to unheeded warnings to rebut the argument that a suit is merely competitive harassment.

Why trade secret cases are hard to litigate

Trade secret disputes turn on questions that are easy to allege and hard to prove. To prevail, Apple will generally need to show three things: that the information at issue qualifies as a trade secret (it derives value from being secret and Apple took reasonable steps to keep it that way), that it was misappropriated (acquired or used through improper means or breach of a duty), and that OpenAI’s products actually rely on that misappropriated information.

The last point is the battleground. Employees move between companies constantly, and they carry general skills and knowledge with them — that mobility is legal and, in California, strongly protected. What is not legal is taking documents, parts, or specific confidential designs. Apple’s allegations about physical parts brought to interviews, a stolen laptop, and downloaded files are attempts to draw that line clearly: not “these people knew how Apple works,” but “these people took Apple’s property.” Whether OpenAI’s forthcoming hardware embodies specific Apple secrets, rather than industry-standard engineering, is the question discovery will chase.

The stakes for both sides

For OpenAI, the timing is awkward. The company is racing to establish a hardware business from scratch while simultaneously scaling its software franchise — it recently repositioned ChatGPT as a work “super app” and has been in unusual discussions with Washington over its corporate structure. A trade secret cloud over the device roadmap complicates fundraising narratives and could, in a worst case, force design changes or injunctive relief on unreleased products.

For Apple, the suit is partly defensive and partly a warning shot. Apple’s competitive edge has always been tight integration of hardware, software, and design, and the departure of senior hardware talent to a well-funded rival is precisely the scenario its confidentiality regime is built to deter. Filing publicly — with language as sharp as “rotten to its core” — signals to current employees that Apple will litigate aggressively when it believes the line has been crossed.

What it means

This is a fight over who gets to build the next general-purpose computing device, dressed as a trade secret complaint. The AI industry has spent two years competing on models; Apple’s lawsuit is a sign that the next front is hardware, and that the incumbent intends to defend that turf in court rather than only in the market.

Who’s exposed: OpenAI and io Products, whose hardware roadmap now carries legal risk before a single product has shipped, and the two named engineers, who face personal liability. Even if OpenAI ultimately prevails, discovery could expose internal design decisions and slow a program that depends on speed.

Who benefits: Apple gains leverage and a deterrent, and rival talent-poachers across the industry will read the complaint as a caution. Litigation of this size also tends to freeze recruiting between the two firms, which advantages whoever already holds the talent.

What to watch: whether OpenAI files to dismiss or narrow the claims, whether Apple seeks an injunction against specific hardware features, and whether the case surfaces concrete evidence — actual documents or parts — versus the softer “knowledge in people’s heads” that courts routinely allow to walk out the door. The gap between those two is the difference between a landmark judgment and a settled distraction. As with Anthropic’s march toward public markets, the AI leaders are discovering that scale brings the scrutiny — legal and financial — that early-stage labs never faced. This time the scrutiny is coming from a former partner, and it is aimed squarely at the future of the device on your desk.

Chisato Chisato · · 6 min read

Lilian Weng Rejoins OpenAI to Lead Self-Improvement

Thinking Machines co-founder Lilian Weng left the startup citing health, then rejoined OpenAI within days to lead a new recursive self-improvement research team.

#AI #OpenAI #Machine Learning