Two years ago Apple built ChatGPT into the iPhone. On July 10, 2026, it asked a federal court to treat OpenAI as a thief. The allegations — unreturned laptops, evaded exit interviews, a coaching playbook for departing engineers — are the most instructive trade secret discovery story in years.
Two years ago, Apple gave OpenAI the most valuable real estate in consumer technology — building ChatGPT directly into the iPhone's operating system. On July 10, 2026, it filed a federal lawsuit in the Northern District of California accusing the same company of orchestrating a systematic campaign to steal Apple's trade secrets, and it asked the court to compel OpenAI to preserve evidence and return confidential materials immediately.
The allegations in Apple's complaint are, by any measure, extraordinary. OpenAI allegedly circulated an internal Apple document bearing a "Need to Know" designation to new hires, detailing how departing Apple employees could avoid the company's security procedures — specifically, how to prevent the "dreaded walkout," which would immediately remove an employee from Apple after giving notice, instead of letting them work their final two weeks and retain access to confidential systems. OpenAI allegedly coached departing Apple employees to let OpenAI know immediately if asked to sign anything at an exit interview, and advised them not to sign.
The former employees named in the lawsuit — Chang Liu, a senior electrical engineer who left for OpenAI in January 2026, and Tang Tan, now OpenAI's Chief Hardware Officer — are accused of downloading dozens of confidential files, failing to return a company-issued laptop, accessing a former colleague's work computer after departure, and bringing Apple hardware components to OpenAI job interviews. Apple alleges OpenAI has hired more than 400 former Apple employees. That number, Apple argues, is not a coincidence. It is a pattern.
The Apple v. OpenAI lawsuit is, on its surface, a trade secret case between two of the most powerful technology companies in the world. But for legal teams and information governance professionals, it is something more specific: a case study in what happens when offboarding procedures fail, when personal device policies are not enforced, and when access controls are not revoked promptly after an employee's departure.
In its August 6 response, OpenAI argued that Apple allowed employees to use personal iCloud accounts for work and failed to properly revoke access after they left the company. It submitted text message records showing that an Apple manager remained logged into the personal iCloud account of defendant Chang Liu after he left the company to transfer files, and later asked him to send additional Apple materials. This is not a defense of OpenAI's conduct — it is a direct argument that Apple's own information governance failures contributed to the alleged theft. And it is exactly the kind of argument that courts take seriously in trade secret litigation.
Apple's complaint, meanwhile, reveals the discovery obligations the case is already generating. Apple asked the court to expedite discovery, saying its internal investigation indicates some additional former employees may have participated in, or witnessed, the alleged theft of trade secrets. That is how trade secret discovery works in practice — one name leads to another, one device leads to another, and the scope of the investigation expands as production reveals what existed. The case is barely six weeks old and Apple is already asking for accelerated discovery timelines.
OpenAI argued that Apple allowed employees to use personal iCloud accounts for work and failed to revoke access after departure. Apple's offboarding failures became OpenAI's defense. That is an information governance lesson, not just a legal one.
The Apple v. OpenAI case is unusual in scale and visibility. The underlying vulnerabilities it exposes are not. Every organization that has employees with access to confidential information faces the same category of risk — and the same category of discovery obligation — when those employees leave and that information ends up somewhere it should not be.
The specific failures Apple alleges — an unreturned laptop, personal cloud accounts used for work, access not revoked at departure, an exit interview not completed — are not exotic security failures. They are among the most common offboarding gaps in organizations of every size and industry. The Apple v. OpenAI case makes those gaps visible because both parties are household names. But the legal exposure those gaps create is identical whether the company is Apple or a mid-sized law firm or a regional healthcare provider.
For legal teams managing litigation or anticipating it, the case also illustrates how quickly digital evidence in trade secret matters proliferates. A single unreturned laptop becomes a forensic examination. A personal iCloud account becomes a production dispute. A text message between a manager and a departed employee becomes evidence of the organization's own security failures. The discovery footprint of a trade secret case is rarely contained to the obvious — it expands into every communication, access log, and device that touched the information at issue.
The Apple v. OpenAI case is the most visible trade secret litigation of 2026. But the discovery vulnerabilities it reveals — unreturned devices, personal cloud accounts used for work, access not revoked at departure, exit interviews not completed — are among the most common information governance gaps we encounter in matters far less prominent than this one.
At Sovereign Discovery, we approach employee departure as a discovery event, not an administrative formality. The window between an employee's resignation and their final day is often the only opportunity to collect devices defensibly, revoke access completely, and create a documented record of what was returned and what was signed. The Apple v. OpenAI case demonstrates what happens when that window closes without those steps being taken. The discovery that follows is more expensive, more expansive, and more damaging than any exit interview would have been.
Apple gave OpenAI a home inside every iPhone. Two years later it is asking a federal court to compel OpenAI to return confidential hardware files, stop using its trade secrets, and preserve evidence across dozens of devices and accounts. The irony is almost too large to fit in a complaint.
But the lesson is not about Apple and OpenAI. It is about the gap between the moment an employee hands in their badge and the moment your organization is certain that all access is revoked, all devices are returned, and all confidentiality agreements are signed. That gap — measured in hours or days — is where trade secret litigation begins. The Apple v. OpenAI case is simply the most expensive version of a story that plays out in organizations of every size, every day.