Apple v. OpenAI Trade Secrets Lawsuit

Apple's complaint doesn't describe a single leaked file - it describes a system. Recruiters allegedly told job candidates who were still employed at Apple to bring actual hardware components, CAD designs, and prototypes into interviews for what the filing calls 'show and tell' sessions ^[1]. Bloomberg's coverage of the case traces that pattern back to OpenAI's roughly $6.5 billion acquisition of Jony Ive's io team, framing Apple's core claim as a 'coordinated' recruiting campaign in which staff departing or interviewing at OpenAI were allegedly told to bring actual parts, drawings, and supplier information for unreleased Apple hardware. Tang Yew Tan, a 24-year Apple veteran who left in February 2024 to become OpenAI's Chief Hardware Officer, is accused of using his own insider knowledge to draw confidential details out of candidates ^[3]. Separately, former systems engineer Chang Liu is accused of exploiting a leftover authentication vulnerability to keep downloading Apple's confidential engineering files after he had already left the company ^[1]. By early August, Apple's internal investigation had widened the case to at least 13 named or implicated former employees ^[4], a small slice of the roughly 400 former Apple employees Apple's complaint says now work at OpenAI in total ^[1][2], and the company has now asked the court for a preliminary injunction and expedited discovery to freeze any use of its information while the case is litigated ^[3].

OpenAI's August 6 motion to dismiss doesn't just deny the allegations - it goes on offense. The 31-page filing argues Apple's complaint fails to identify a protectable trade secret or plausibly allege misappropriation, and accuses Apple of trying to paper over its own competitive weaknesses: 'Apple should not be permitted to use a baseless and pretextual lawsuit to make up for its shortcomings in the market for talent and retaining its employees, and its failures to integrate AI into its products' ^[5]. The rhetoric is relentless throughout the filing: across those 31 pages, variations of the word 'fail' appear nearly 50 times ^[5]. OpenAI's specific counter is that Apple's own security practices - including letting staff use personal iCloud accounts for work - are what left former employees with leftover system access, not evidence of a coordinated theft ring ^[6]. That dueling framing has split public reaction along similar lines: Reddit's read on the case has skewed heavily toward crediting Apple's version, even inside OpenAI's own community, while the highest-engagement post on X - a widely shared report from Bloomberg's Mark Gurman - simply relayed the motion's own language, quoting OpenAI's characterization of the suit as a 'baseless and pretextual lawsuit' and carrying that framing to an audience well beyond the court filing itself. Beyond the courtroom, that dueling narrative carries its own cost: with Apple alleging systemic, leadership-endorsed misconduct and OpenAI countering that Apple's own lax security and talent retention are to blame, both companies risk reputational and recruiting fallout that could dent trust with employees, recruits, and partners on either side ^[5].

The lawsuit only makes sense against the backdrop of OpenAI's hardware push. OpenAI's $6.5 billion acquisition of Jony Ive's design firm io signaled a serious move into consumer AI devices, and it's specifically implicated in Apple's claims that OpenAI drew on Apple's manufacturing and battery design knowledge ^[1]. Notably, Ive himself is not named as a defendant, a gap that has drawn its own scrutiny given how central the io deal is to the underlying dispute ^[7]. Read together with Apple's own pivot away from OpenAI - ending its ChatGPT-in-Siri arrangement in favor of a custom Gemini-powered Siri built with Google ^[8]- the lawsuit looks less like an isolated legal dispute and more like the fallout from two companies' hardware and AI roadmaps colliding head-on.

The immediate practical question isn't whether Apple ultimately wins the case - it's whether Apple can get a preliminary injunction before then. Apple's motion, filed August 3-4, asks the court to bar OpenAI and the named former employees from using or acquiring Apple's confidential information while the case proceeds ^[3]. A hearing on that motion is set for October 1, 2026, with OpenAI's formal response due August 17 ^[5]. If granted, an injunction could delay or constrain OpenAI's consumer hardware development directly; if denied, it would hand OpenAI an early momentum win and undercut the urgency behind Apple's 'irreparable harm' argument. Either outcome will likely shape how aggressively both companies litigate the rest of the case.