OpenAI targets smartphones with Qualcomm

Smartphones are the next obvious battleground for AI — not because the hardware is exciting, but because the phone is still the thing people use all day. That is why the latest OpenAI rumor matters. Ming-Chi Kuo, the Apple supply-chain analyst whose notes move markets, says OpenAI is working with Qualcomm, MediaTek, and Luxshare on an AI-first smartphone that could enter mass production in 2028. Qualcomm stock jumped on the report, which tells you investors think this is more than random gossip. (cnbc.com) ### What is the actual claim? The claim is pretty specific. Kuo says Qualcomm and MediaTek would co-develop the phone’s processors, while Luxshare would be the exclusive system co-design and manufacturing partner. He also says the project’s final specs and supplier choices could be locked by late 2026 or early 2027, with mass production targeted for 2028. That is much more concrete than the usual “OpenAI might make hardware someday” chatter. (mobileworldlive.com) ### Why does Qualcomm matter here? Because this rumor is not really about a handset first — it is about where the AI runs. Qualcomm already sells the kind of mobile chips that handle on-device AI, and MediaTek plays the same game at huge scale. If OpenAI wants a phone that feels like an assistant first and an app launcher second, chip design i(mobileworldlive.com)t is why the processor names are the most believable part of the story. (cnbc.com) ### Is this the Jony Ive device? Maybe — but that part is still fuzzy. OpenAI officially announced in May 2025 that Jony Ive’s io team would merge with OpenAI, with Ive taking deep design responsibilities across the company’s hardware work. So there is clearly a real hardware program inside OpenAI now. But earlier reporting around io pointed to strang(cnbc.com)ort, not necessarily the whole thing. (openai.com) ### What does “AI-first smartphone” actually mean? Basically, the pitch is that the assistant becomes the operating model. Instead of opening Uber, Spotify, Calendar, and email one by one, you tell the system what you want and it handles the steps. Kuo’s framing — echoed in follow-up coverage — is that AI agents could replace a lot of app-hopping. That does not mean apps vanish overnight. I(openai.com)delegating tasks. (techcrunch.com) ### Why are people skeptical? Because we have seen this movie before. Humane and Rabbit both sold the idea that AI could replace conventional app flows, and both ran into the same wall — reliability. Agents are impressive in demos, but everyday phone use is messy, stateful, and full of edge cases. If OpenAI is really trying this, the hard pa(techcrunch.com)ine life admin. That is an inference, but it follows directly from how agent-style products have struggled so far. (techcrunch.com) ### So is this confirmed? No. Right now, the core report traces back to Kuo and the supply-chain ecosystem around him. OpenAI has not publicly announced a phone, and Qualcomm has not published a formal product reveal tied to OpenAI on its own site. So the right posture is: plausible, specific, but unconfirmed. The rumor got traction because (techcrunch.com)unds like a real manufacturing plan, not fan fiction. (mobileworldlive.com) ### What is the bottom line? The interesting part is not whether OpenAI ships a phone in 2028. The interesting part is that the company now looks serious about owning the interface, not just the model. If that happens, Apple and Google do not just face another AI app. They face a bid to redefine what a smartphone is for. (mobileworldlive.com)