Apple Just Sued OpenAI for Trade Secret Theft — The AI Partnership Era Just Shattered. Here's What That Means for Every No-Code Builder

Yesterday, Apple filed a federal lawsuit against OpenAI alleging systematic trade secret theft. This is not a patent squabble over rounded corners or swipe-to-unlock. Apple claims OpenAI poached its veteran engineers, coached them to evade internal security protocols, and used stolen confidential hardware information to build a competing consumer device business. The same OpenAI that, two years ago, Apple anointed as its marquee AI partner at WWDC 2024. The partnership era just shattered.

If you build software with AI under the bonnet, this is not someone else's problem. The model layer is now in open warfare. Partners are becoming adversaries in real time. And your stack's exposure to any single AI provider just became a question worth asking out loud.

---

TL;DR

• What happened: On July 10, 2026, Apple sued OpenAI in federal court in Northern California for trade secret misappropriation and breach of contract.
• The defendants: OpenAI; Tang Yew Tan, OpenAI's chief hardware officer (24 years at Apple, former VP of iPhone product design); Chang Liu, a former Apple senior electrical engineer; and io Products, a hardware design firm co-founded by Tan that OpenAI acquired.
• The allegation: A "coordinated pattern of misconduct at an institutional level" to solicit and extract Apple trade secrets for OpenAI's own hardware ambitions.
• What Apple wants: An injunction plus return and destruction of all stolen materials.
• Why it matters for no-code builders: OpenAI is not just the LLM powering your app. It's a company that just got sued by its biggest distribution partner for allegedly stealing secrets to compete with them. That changes the risk calculus for anyone building on OpenAI's stack.

---

What exactly is Apple accusing OpenAI of doing?

The complaint names three defendants alongside OpenAI itself. Tang Yew Tan spent nearly a quarter century at Apple, most recently as vice president of product design for iPhone and Apple Watch. He left Apple, co-founded a hardware design firm called io Products, and then OpenAI acquired that firm and installed Tan as its chief hardware officer.

Chang Liu is a former senior system electrical engineer at Apple. The lawsuit accuses him of taking confidential information about unreleased Apple products and bringing it to OpenAI.

Apple's core claim is that this was not rogue behaviour from a couple of disgruntled ex-employees. It was, in Apple's words, a "coordinated pattern of misconduct at an institutional level." The filing alleges that OpenAI actively coached former Apple employees on how to evade Apple's security protocols while extracting proprietary hardware information.

This is a serious federal lawsuit, not a PR stunt. Apple is seeking an injunction to stop OpenAI from using any of the allegedly stolen information, plus the return and destruction of all materials. If Apple prevails, OpenAI's hardware division could be gutted.

What hardware is OpenAI actually building?

OpenAI has been quietly building a consumer hardware operation. Through io Products, it's developing devices that would compete directly with Apple's iPhone and wearable ecosystem. OpenAI recently unveiled Jalapeño, its first custom silicon chip, built in partnership with Broadcom. The chip is designed for LLM inference workloads. It went from concept to lab testing in nine months.

That speed is part of why Apple is suspicious. Apple knows how long hardware R&D takes. When a competitor ships a custom chip in nine months after hiring two dozen of your senior engineers, you start asking questions about where the shortcuts came from.

OpenAI is no longer just a model company. It wants to own the device, the chip, and the cloud. That puts it on a collision course with Apple, Google, and Samsung simultaneously.

The partnership is now radioactive

In June 2024, Apple stood on stage at WWDC and announced ChatGPT integration into Apple Intelligence. Apple would handle on-device privacy and the user experience. OpenAI would power the cloud intelligence layer. Both companies stood to gain enormously.

That partnership now looks like a historical artefact. Apple is suing its marquee AI partner for allegedly stealing trade secrets to build competing hardware. The ChatGPT integration deal gives OpenAI access to billions of Apple user prompts. Apple's lawsuit does not explicitly address whether that data pipeline is affected, but the trust required to sustain such an arrangement has plainly evaporated.

Whether the integration survives in its current form is an open question. Apple has already started hedging by exploring Google Gemini as an alternative provider for Siri. The multi-provider pivot was already underway. This lawsuit accelerates it.

This fits a pattern. The model layer is destabilising.

If you have been reading nocode.tech this year, you have watched this destabilisation happen in real time.

In March, the US government bought a 5% stake in OpenAI, raising uncomfortable questions about whose interests the company serves. In June, the US government imposed export controls that shut down Anthropic's most powerful models, Claude Fable 5 and Mythos 5, outside American borders. Then, just this week, OpenAI released GPT-5.6 with usage gates and geographic restrictions that make it clear the model layer is now a geopolitical chessboard, not a neutral utility.

And now Apple, OpenAI's biggest distribution partner, is suing it for trade secret theft.

Each of these events, on its own, is manageable. Stacked together over four months, they tell a story: the AI model layer is fragmenting into armed camps. The era of friendly API calls between partners is giving way to lawsuits, export controls, and government ownership stakes.

What this means for no-code builders

If your app runs on OpenAI's API, or you built agents using ChatGPT Work, or you rely on Bubble's AI Agent system, your vendor risk profile just changed. Not because OpenAI is going to disappear overnight. It won't. But because the relationship between the companies that build AI models and the companies that distribute them has become adversarial.

When partners become litigants, platform stability suffers. Roadmaps get delayed. API terms get renegotiated. Integration pipelines get paused while lawyers review data-sharing agreements. None of this is hypothetical. These are the predictable second-order effects of a federal trade secret lawsuit between major corporate partners.

This is also where the no-code argument for platform abstraction gets real. The tools that let you swap models without rewriting your stack, the platforms that sit above the LLM layer rather than inside it, the builders that treat AI as a configurable capability rather than a dependency on one company's API, those are the ones that hold up when the model layer catches fire.

Structured no-code platforms that abstract the AI layer are not just convenient. In this environment, they are the safe harbour.

The takeaway

The Apple-OpenAI lawsuit is not just drama between two of the most valuable companies on earth. It is a signal that the AI industry's partnership model, the one that made "ChatGPT inside every Apple device" seem inevitable, is breaking down.

For no-code builders, the lesson is straightforward. The model layer is unstable by design. It will keep fragmenting. Lawsuits, export controls, and government interventions will keep piling up. Building your product on top of a single model provider is a bet that the partnership era holds together. That bet just got a lot worse.

Build on platforms that let you switch models when the ground shifts. Because the ground is shifting faster than anyone expected.