Two years after embedding ChatGPT in the iPhone, Apple is asking a federal court to stop OpenAI from using its hardware secrets to build a competing device.
Two years after embedding ChatGPT in the iPhone, Apple is asking a federal court to stop OpenAI from using its hardware secrets to build a competing device.

Two years after embedding ChatGPT in the iPhone, Apple is asking a federal court to stop OpenAI from using its hardware secrets to build a competing device.
Apple filed a federal lawsuit accusing OpenAI of stealing proprietary hardware designs and supplier data, escalating a conflict between the iPhone maker and its former AI partner over the next generation of consumer devices.
"The misconduct ran at every level, from members of its Technical Staff to its Chief Hardware Officer, and in coordination with business partners," Apple said in the complaint filed in the Northern District of California.
The suit names Tang Yew Tan, a 24-year Apple veteran who oversaw iPhone and Apple Watch product design before becoming OpenAI's chief hardware officer, and Chang Liu, a former Apple engineer. Apple alleges Tan forwarded supplier information to his personal email and directed job candidates to bring unreleased Apple hardware parts — including batteries, logic boards, and System-in-Package modules — to interviews. Liu is accused of downloading dozens of confidential technical documents after leaving and keeping an Apple-issued laptop. OpenAI acquired io Products, the hardware studio co-founded by former Apple design chief Jony Ive and Tan, for roughly $6.5 billion in 2025.
The lawsuit lands as OpenAI prepares to debut its first consumer device — a screenless smart speaker designed as an AI-powered home companion — before the end of this year, with a broader market release planned for 2027. Apple's complaint, which seeks an injunction barring OpenAI from using the alleged secrets, threatens to delay that timeline and shows how seriously the iPhone maker views the competitive threat.
What the Allegations Reveal About OpenAI's Hardware Plans
Every detail Apple chose to highlight sketches a piece of the device OpenAI has never formally announced. System-in-Package modules are how engineers fit a computer into something the size of a hearing aid. Supplier relationships are what a company needs when a prototype becomes a production line. A proprietary metal-finishing technique matters to a team deciding how mass-manufactured hardware feels in a customer's hand. Apple, in the act of protecting its secrets, has confirmed which engineering challenges a consumer AI device actually requires.
OpenAI's chief global affairs officer told Axios in January that the company aims to debut its first device before the end of 2026. Subsequent reporting describes a screenless wearable and a smart speaker with a camera, with parts of the lineup slipping toward early 2027. Apple's own filing treats the program as far along — a company does not ask a federal court for relief against a product it believes will never ship.
The Talent Bottleneck Behind the Lawsuit
OpenAI has shown it can buy nearly every input it needs. Compute arrives through multiyear cloud commitments, capital through the largest private raises on record, distribution through a chatbot used by hundreds of millions. Consumer hardware runs on a scarcer input: engineers who have shipped miniaturized electronics at hundred-million-unit scale, and the overwhelming majority of them trained at one company in Cupertino.
California famously does not enforce noncompete agreements, which is why trade secrets law is the only brake Apple has on the movement of its own alumni. Apple says more than 400 former employees now work at OpenAI. The lawsuit is what a talent bottleneck looks like at the moment it finally binds.
What the Timing Means for Investors
The suit lands in the middle of a leadership handover at Apple. Tim Cook becomes executive chairman on Sept. 1 and hands the chief executive role to John Ternus, the hardware engineering chief who spent his career inside the organization the complaint defends. An Apple about to be run by its top hardware engineer, suing to protect hardware secrets, is a consistent message about where the company believes its franchise is won or lost.
For OpenAI, a first-party device is how it stops renting its distribution. Today ChatGPT reaches consumers through hardware other companies control — the iPhone first among them — and every one of those channels can reprice or restrict it. That dependency is what a $6.5 billion acquisition and a recruiting campaign reaching into Apple's senior ranks were built to end.
The case will grind through motions for months. An early injunction ruling would show whether OpenAI's timeline is genuinely at risk. A device unveiling in the coming months would show the program outran the lawsuit. Either way, the filing has already confirmed what both companies agree on: the next major consumer computing platform is being designed right now, and the fight over who ships it has moved from the labor market into federal court.
This article is for informational purposes only and does not constitute investment advice.