OpenAI has fired back at Apple's trade secret lawsuit by publishing chat logs and emails that appear to show Apple employees voluntarily contacting a former colleague for technical help after he left the company. The AI research and deployment company also accused Apple of sloppy legal tactics, including sending an initial email to the wrong person and claiming a phone call that never happened.
Apple sued OpenAI in a federal court in California, alleging systematic theft of trade secrets. The complaint centers on former Apple engineer Chang Liu, who is accused of improperly accessing confidential information after joining OpenAI. Apple further claims that OpenAI actively encouraged departing employees to steal files and specifications during their notice periods.
OpenAI's Response: A Blog Post and a Paper Trail
OpenAI published a blog post accusing Apple of taking a "sloppy and unnecessarily aggressive approach" to the dispute. To back that claim, the company released iMessage threads and email correspondence involving Liu, whose last day at Apple was January 22, 2026.
Just five days later, on January 27, 2026, an Apple employee reached out to Liu for a technical assessment. The message read: "Of course, I could ask several folks, but you are the best. Even if you don't work here anymore." Liu responded with details about internal Apple decisions.
The same employee contacted Liu again on February 14, 2026, this time asking about schematics. Then, on March 5, 2026, Liu was added to a group chat with several Apple employees. In that chat, he directed them to internal folders and contacts. Liu ended the conversation himself, writing: "Hi, this is highly irregular, please remove me from this thread."
OpenAI argues that this "residual access" is a well-known problem at Apple, stemming from poor access management. The company suggests the messages show Apple's own staff relied on Liu even after his departure, undermining the narrative that he was the one improperly seeking information.
Legal Missteps by Apple's Lawyers
OpenAI also detailed a series of errors in Apple's initial outreach. Apple claimed it contacted OpenAI as early as February without receiving a response. But OpenAI says the email went to the wrong person because Apple's lawyers mixed up two Asian last names. Apple's outside lawyer also claimed to have spoken with OpenAI's General Counsel by phone, a call OpenAI says never happened.
Apple only acknowledged both errors after OpenAI pointed them out. The specific allegations in the lawsuit were never raised in earlier correspondence. Instead, Apple's lawyer said the company was working to "sort out any issues." Five months of silence followed before Apple filed suit.
Stay ahead of the AI curve
The most important updates, news, and content — delivered weekly.
No spam. Unsubscribe anytime.
The Case Against Tang Tan
The lawsuit also names Tang Tan, a former Apple executive who spent more than 24 years at the company and was considered one of its most innovative leaders. OpenAI rejects the allegations against him, claiming Tan consistently told his team that proprietary information from other companies was neither welcome nor to be used.
The lawsuit is tied to OpenAI's growing hardware ambitions, specifically the startup io Products, co-founded by Jony Ive, Apple's iconic former designer. That connection adds weight to Apple's concerns about talent and technology moving to a rival.
What the Evidence Does and Doesn't Show
The chat messages and email blunders make for compelling reading, but they don't refute Apple's core accusations. Liu's iMessage threads may show Apple had its own access management problems, but they don't address the broader claim that OpenAI encouraged new hires to bring proprietary information.
According to the complaint, more than 400 former Apple employees now work at OpenAI. That figure gives the lawsuit a wide scope, even if the specific chats released by OpenAI paint a more sympathetic picture of Liu's conduct.
The article, written by Matthias Bastian for The Decoder and published on August 4, 2026, notes that OpenAI's defense may be partially successful in highlighting Apple's sloppiness. But the core allegations remain unresolved. The case will likely hinge on whether OpenAI can show a pattern of proper conduct, or whether Apple can prove a systematic effort to harvest its secrets.
For now, the public record shows a former engineer who was repeatedly sought out by his old colleagues, and a legal team that stumbled through its own outreach. That may not win the case for OpenAI, but it has certainly complicated Apple's story.