-
\t
- Qualcomm surged 13% on a report that it’s partnering with OpenAI to build an AI-first smartphone.
- The device targets 300-400 million annual shipments by 2028, exceeding iPhone volumes.
- OpenAI’s phone would replace apps with AI agents, requiring full control of hardware and software.
\t
\t
Qualcomm shares jumped 13% on Monday after TF International Securities analyst Ming-Chi Kuo reported that the chipmaker is partnering with OpenAI to develop a custom smartphone processor. The partnership includes MediaTek, which will jointly design the chip, while Luxshare Precision Industry will exclusively manufacture the device. Kuo projects 300 to 400 million annual shipments if the device succeeds, a figure that would exceed Apple’s iPhone volumes and place the phone in direct competition with the two companies that control roughly 40% of the global smartphone market.
The supply chain is credible—Luxshare builds AirPods for Apple, and Qualcomm powers 75% of Galaxy S26 units. But OpenAI has never shipped hardware, and every previous AI device has failed. The Humane Pin and Rabbit R1 both launched to scathing reviews and were discontinued within months. This is OpenAI’s second hardware track alongside the Jony Ive project, which Altman has said would not be a smartphone.
Kuo explained why OpenAI would make a phone despite the graveyard of failed AI devices. “Only by fully controlling both the operating system and hardware can OpenAI deliver a comprehensive AI agent service,” wrote Kuo on X. “The smartphone is the only device that captures the user’s full real-time state, which is the most important input for real-time AI agent inference.”
Why OpenAI Needs Its Own Phone
OpenAI’s smartphone would replace apps with AI agents that handle tasks directly. Instead of opening a flight booking app, you’d tell the phone to book a flight. Instead of opening a spreadsheet, you’d ask it to compile market data. Kuo sketched out a concept where the homescreen is replaced by a panel of information and ongoing agentic AI tasks. This requires both on-device edge intelligence and cloud AI integration, which is why OpenAI needs custom silicon from Qualcomm and MediaTek.
The timing is aggressive. Specifications and the supplier list are expected to be finalized by late 2026 or the first quarter of 2027, with mass production targeted for 2028. That gives OpenAI less than two years to design, test, and manufacture a device that competes with the iPhone. Qualcomm and MediaTek have the technical expertise—most premium Android phones in 2026 use either Qualcomm’s Snapdragon 8 Elite Gen 5 or MediaTek’s Dimensity 9500 chips. But OpenAI has no track record in hardware manufacturing.
The revenue model is unclear. Kuo suggested OpenAI might tie hardware sales to subscription plans and cultivate a developer ecosystem around AI agents. But that model is unproven. Apple sells hardware at premium prices and takes a 30% cut of App Store revenue. OpenAI would need to convince users to pay for both the device and a subscription to its AI services, all while competing with free alternatives from Google and Apple.
The Hardware Graveyard OpenAI Must Avoid
Every AI-first device launched so far has failed. The Humane Pin raised $230 million from investors including OpenAI’s Sam Altman, launched at $699, and was discontinued less than a year later. The Rabbit R1 raised $30 million on Kickstarter, launched at $199, and was widely panned for being slow, unreliable, and redundant. Both devices promised to replace smartphones with AI assistants, but neither could match the convenience of a phone with apps.
OpenAI’s advantage is its AI technology. GPT-5.5 is the most capable language model available, and OpenAI has access to more compute and data than any startup. But that doesn’t translate to hardware expertise. Designing a smartphone requires managing supply chains, negotiating with component suppliers, and passing regulatory certifications. Apple has been doing this for 15 years. OpenAI has never done it.
The 300-400 million shipment target is also ambitious. Apple shipped 235 million iPhones in 2025. Samsung shipped 227 million smartphones. OpenAI would need to capture roughly 20% of the global smartphone market to hit the low end of Kuo’s projection. That would require convincing hundreds of millions of users to abandon their iPhones and Android phones for an unproven device from a company with no hardware track record.
Qualcomm’s stock surge reflects investor optimism about the partnership. The chipmaker has been pushing into AI data center chips and won a contract to ship to “a large hyperscaler” within the calendar year, according to CEO Cristiano Amon. But the OpenAI smartphone is still a rumor from an analyst, not an announced product. Mass production is two years away, and a lot can change between now and 2028.
Kuo’s report was posted on X on April 26, 2026. The Next Web and Wccftech first reported the details.
