We may need to make room for a third global smartphone and OS (operating system), besides Google Android and Apple iPhone after all. New media reports continue to heighten expectations for an OpenAI Smartphone next year.
I’d referred to this possibility on this AI-RTZ substack and ARD podcast for a few days now.
A key difference here is OpenAI seems to be planning this device complete with its own operating system. So not just drafting off Google’s Android like Samsung and countless Chinese OEMs have done around the world. And Apple’s wholly controlled iOS operating system, tightly linked to the other Apple OSes across its platforms.
“Latest industry checks: OpenAI is working with MediaTek and Qualcomm to develop smartphone processors, with Luxshare as the exclusive system co-design and manufacturing partner. Mass production is expected in 2028.”
• “AI Agent redefines the smartphone: ”Users are not trying to use a pile of apps. They are trying to get tasks done and fulfill needs through the phone. This fundamentally changes how people think about smartphones. I made a smartphone interface concept design, shown at the end of this post, for comparison with today’s model, using the iPhone as an example.”
Then what in my view, is the likely $100+ billion question for OpenAI, which will consume at least 25% of ALL its resources over the next 3+ years. A ‘side quest’ that may be a major mission in its own right.
• “Why would OpenAI make a phone? “
Sam , Jony and team have likely convinced themselves of the following:
”1. Only by fully controlling both the operating system and hardware can OpenAI deliver a comprehensive AI agent service.”
This has both strategic and technical truths. Meta founder/CEO Mark Zuckerberg found this out the hard way in his Metaverse and other ambitions, delivered via regular apps on iphone and Android devices. Core hardware and software integration with the devices is a must. Thus the tens of billions being expended by Meta on its AI smart glasses ambitions.
Then the reality of AI agentic operations controlling all the sensors on a smartphone continuously to capture data. Oh, and of course serve the user’s daily needs, with personalized memories and data stored on the device and cloud as needed.
”2. 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.”
Not only for user applications, but to train the broader LLM and SLM AI models on the device and in the cloud. This is the critical Box 4 of the AI Tech Stack that I’ve talked about for a long time. It’s what made Google Google. Billions of user queries every day making the core search service smarter and better. Not to mention fueling ad dollars.
Then it’s the ultimate recognition by the biggest AI companies trying to figure out the form factor to deliver AI services. They may not be smart glasses, wearables, and other doodads on the mainstream user’s person.
It’s likely the tried and true smartphone.
Already in billions of user hands, being obsessively used daily for hours on end. More than any other device in those mainstream users’ lives. Daily habits baked and compounding in every facet of their lives.
A market bigger than the AI business today with 3+ hundred million devices sold every year.
”3. Smartphones will remain the largest-scale device category for the foreseeable future.”
“• Tightly integrated cloud and on-device AI:” ”1. The phone needs to continuously understand the user’s context. Power consumption, memory hierarchy management, and basic small-model execution will be key processor design considerations.”
Handing off larger context window token queries to the cloud off the phone when needed. Again, with deep integration of hardware and software that Apple and Google will not allow third parties, with their operating systems.
”2. More complex or compute-intensive tasks will be handled by cloud AI.”
“• OpenAI’s advantages lie in its consumer brand, years of accumulated user data, and leading AI models. Smartphone hardware is already highly mature, so OpenAI can work with the supply chain to develop the device. On the business model side, OpenAI may bundle subscriptions with hardware and build a new AI agent ecosystem with developers.”
OpenAI seems to have made headway on the critical component and assembling partners. They are all world-class companies already close suppliers to Apple and Android world OEMs with billions of dollars at stake.
”• MediaTek and Qualcomm are processor co-development partners and could benefit from long-term replacement demand:” ”1. Specifications and suppliers are expected to be finalized by late 2026 or 1Q27.” ”2. Taking MediaTek × Google TPU Zebrafish as an example, the revenue contribution of a single chip is roughly equivalent to 30–40 AI agent smartphone processors. If the initial target is the global high-end smartphone segment, which ships about 300–400 million units per year, the replacement cycle could become another major growth driver.”
“• No matter how hard Luxshare tries, it will be difficult for the company to surpass Hon Hai’s assembly position in Apple’s supply chain. That makes this project especially meaningful for Luxshare. With an early position in the supply chain, Luxshare could become a leading beneficiary in the next smartphone generation.”
“[Industry Check Update】OpenAI appears to be fast-tracking its first AI agent phone, with mass production targeted as early as 1H27. Potential drivers include supporting a year-end IPO narrative and intensifying competition in AI agent phones. MediaTek currently appears better positioned to become the sole processor supplier, with the device set to use a customized version of the Dimensity 9600, built on TSMC’s N2P node in 2H26. The ISP is the headline spec, with an enhanced HDR pipeline improving real-world visual sensing. Other key specs include a dual-NPU architecture for heterogeneous AI compute, LPDDR6 + UFS 5.0 to ease memory bottlenecks, and pKVM + inline hashing for security. If development stays on track, combined 2027–2028 shipments could reach around 30 million units.”
Esecially given controllable and uncontrollable challenges involved in doing a global smartphone from scratch.
There is a reason why even the most formidable tech companies, Microsoft, Amazon,Meta and others have not yet cracked this challenge. And Google with Pixel smartphones is atthe periphery of smartphones around the world.
Despite all that, as a gadget nerd and self-professed geek, I am excited that OpenAI is giving this a go. Having used AI Agents services intensively in recent months, it’s increasingly obvious that not only smartphones, but computer and laptop operating systems need to be re-thought for this AI Tech Wave going forward.
“AI makers have trained agents to use software like people do. Now the industry is starting to build software meant for agents instead.”
“Why it matters: If agents stop using software the way humans do, tech competition could shift from who has the best interface to who controls the APIs, data and permissions agents need to act.”
“The big picture: Today’s agents are like self-driving cars forced to navigate roads built for humans — stoplights, signage and all.”
Using Anthropic’s Claude CoWork daily on my Mac Mini is an exercise of sometimes frustrating SHARING of the computer while the AI Agents try to do their own thing using all the resources of the computer. While you’re trying to edit a document, or read a web page.
We’re likely heading into a world where early adopters initially will likely opt for separate computers and smartphones for their AI tasks indepdent of their personal and professional work.
Also, it’s clear that the two major smartphone operating system companies each have their ecosystem and incumbency issues that will be slower to innovate around AI on ‘first principles’ relative to a new AI smartphone startup.
Google does not control its hardware destiny with dozens of OEMs doing their own UI/UX layters and applications alongside of Android. Especially in Asian markets like China and beyond. Note how Google and Samsung, its principal OEM in the US market have had dueling and duplicative messaging and other first party apps on Samsung devices for years.
In that world, it becomes clearer how OpenAI and team are getting the conviction to do their own AI smartphone. Despite Apple and Google smartphone world dominance.
As long as they’re going in eyes wide open. And communicating clearly with all their stakeholders around the efforts. It’s a promising move in this AI Tech Wave. But with big price tags and challenges. Stay tuned.
(NOTE: The discussions here are for information purposes only, and not meant as investment advice at any time. Thanks for joining us here)
