The iPhone is not actually being disrupted by AI at all
In a moment that signals both the ambition and the anxiety of the artificial intelligence era, OpenAI is reportedly preparing to enter the hardware arena with a smartphone built not around apps, but around AI agents — autonomous systems designed to act on a user's behalf. The move represents a philosophical break from how we have understood personal computing for decades, and a quiet admission that integrating intelligence into existing devices may not be enough. Whether this is the beginning of a new paradigm or a costly overreach remains the defining question of the endeavor.
- OpenAI is reportedly partnering with MediaTek, Qualcomm, and Luxshare to bring an AI-native smartphone to mass production by 2028 — a striking reversal from CEO Sam Altman's long-held skepticism about proprietary hardware.
- The core tension is philosophical: rather than adding AI features to an app-based device, OpenAI wants AI agents to replace the app model entirely, a bet that users are ready to interact with their phones in a fundamentally different way.
- Altman's public call to 'rethink operating systems and user interfaces' signals deep frustration with how slowly AI has actually changed the smartphone experience, even as Apple and others have layered on AI features.
- Critics are swift and pointed — observers note that the vast majority of smartphone use involves gaming, streaming, and social media, none of which an AI-agent-centric device is positioned to serve well.
- The project faces not a technical wall but a behavioral one: billions of users are deeply embedded in iOS and Android ecosystems, and no vision of computing, however elegant, has yet dislodged that loyalty.
OpenAI is building a smartphone — a sentence that would have seemed unlikely just months ago, given CEO Sam Altman's repeated resistance to the idea that his company needed its own hardware. But analyst Ming-Chi Kuo of TF International Securities reported on April 26 that OpenAI is partnering with chipmakers MediaTek and Qualcomm, alongside manufacturer Luxshare, with a target of mass production by 2028.
The device is not conceived as a conventional iPhone rival. At its core would be AI agents — autonomous systems capable of handling tasks on a user's behalf — replacing the familiar model of discrete apps. Instead of opening an email client or a browser, a user would simply interact with an intelligent layer designed to manage those needs. It is a vision of computing that is more fluid, more anticipatory, and far less tested at scale.
The timing carries meaning. On the same day Kuo's report appeared, Altman posted publicly that it 'feels like a good time to seriously rethink how operating systems and user interfaces are designed.' Perplexity CEO Aravind Srinivas had recently observed that the iPhone, despite its AI additions, remains fundamentally an app-based device — not truly disrupted. OpenAI's apparent conclusion is that disruption requires building the hardware from the ground up.
Skepticism arrived quickly. Critics noted that the overwhelming majority of smartphone use — gaming, streaming, social media — falls outside what an AI-agent experience is built to deliver. One observer called the concept 'dead on arrival,' pointing to the entrenched ecosystems of iOS and Android that users have no obvious reason to abandon.
The real obstacle OpenAI faces is not engineering but habit. Smartphones are so woven into daily life that a new device must match everything users already have while offering something they didn't know they needed. Whether an AI-native phone can clear that bar, or whether it becomes a cautionary lesson in the distance between vision and adoption, is the question that will follow this project for years.
OpenAI is building a smartphone. That sentence alone marks a significant reversal for a company whose CEO, Sam Altman, has long resisted the idea that his artificial intelligence firm needed to make its own hardware to compete with Apple. But according to analyst Ming-Chi Kuo of TF International Securities, the calculus has shifted. In a report published on April 26, Kuo detailed plans for OpenAI to partner with chipmakers MediaTek and Qualcomm, along with manufacturer Luxshare, with the goal of bringing a device to mass production by 2028.
The phone would not be another iPhone competitor in the traditional sense. Instead, Kuo's reporting suggests that AI agents—autonomous systems capable of performing tasks on behalf of the user—would form the foundation of the entire experience. This represents a fundamental departure from how smartphones work today, where users navigate between discrete applications. The distinction matters. Rather than opening an app to check email or search the web, a user would interact with an intelligent agent designed to handle those tasks. It is, in theory, a more seamless and intuitive way to use a device.
The timing of this pivot is telling. On the same day Kuo published his report, Altman posted on X that it "feels like a good time to seriously rethink how operating systems and user interfaces are designed." The message hints at frustration—a sense that current smartphones, even those running the latest software, have not truly integrated AI in a way that fundamentally changes how people interact with their devices. Perplexity CEO Aravind Srinivas recently made a blunt observation: the iPhone, despite all the AI features Apple has added, is not actually being disrupted by artificial intelligence. It remains, at its core, an app-based device. OpenAI's move suggests the company believes the only way to change that dynamic is to build the hardware itself.
But the skepticism is immediate and substantial. On social media, critics pointed out the obvious gap in OpenAI's vision. A phone built around AI agents and ChatGPT integration would struggle to do what billions of people actually use their phones for: play games, stream video, scroll through social media. One commenter captured the tension succinctly: "AI can't play games, watch Netflix, and scroll social media. That is why 99% have phones." Another predicted the device would be "dead on arrival," unable to compete with iOS and Android ecosystems that have spent years building out the infrastructure users depend on.
The challenge OpenAI faces is not technical but behavioral. Smartphones have become so embedded in daily life that any new device must not only match their capabilities but exceed them in ways users actually want. An AI-centric phone might excel at certain tasks—research, writing, problem-solving—but it would need to do everything else just as well. Whether consumers will abandon devices they know and trust for a vision of computing that remains largely theoretical is the question that will determine whether this project succeeds or becomes another cautionary tale about the limits of disruption.
Citações Notáveis
It feels like a good time to seriously rethink how operating systems and user interfaces are designed— Sam Altman, OpenAI CEO, on X
The iPhone is actually not getting disrupted by AI at all— Aravind Srinivas, Perplexity CEO
A Conversa do Hearth Outra perspectiva sobre a história
So OpenAI is making a phone now. Why? Sam Altman used to say they didn't need to.
He did. But the frustration seems real—current phones haven't actually changed how we use them, even with AI built in. They're still app-based. Altman's saying it's time to rethink that.
And the answer is to make their own device?
That's the theory. If AI agents are the core experience instead of apps, the phone works differently. You don't open email—the agent handles it. You don't search—the agent knows what you need.
That sounds elegant. So why are people so skeptical?
Because people use phones for things AI agents can't do well yet. Gaming, streaming, social media. Those aren't edge cases—they're why most people have phones.
Is this a real product or a strategic bluff?
The partnerships with MediaTek and Qualcomm suggest it's real. But real and viable are different things. The skeptics might be right—you can't replace a device people depend on with something that only does some things better.
What would it take to work?
It would have to do everything the current phones do, plus the AI things better. That's a much higher bar than just being different.