AI Agents Will Replace Smartphone Apps: Nothing CEO Carl Pei’s Revolutionary Vision for a Post-App Future

In a bold prediction that challenges the foundation of modern mobile computing, Nothing CEO Carl Pei declares that smartphone applications are headed for obsolescence, destined to be replaced by proactive artificial intelligence agents. Speaking at the SXSW conference in Austin, Texas, on Wednesday, the tech visionary outlined a future where our devices anticipate needs and execute tasks seamlessly, rendering today’s app-centric model archaic. This perspective comes as his company, known for its transparent-design phones, secured significant funding to pursue this very AI-first ambition.

The Inevitable Disruption of Smartphone Apps

Carl Pei, co-founder of the consumer electronics brand Nothing, presents a stark warning for the software industry. He argues that the core value proposition of most startups—their proprietary application—faces imminent disruption. “In terms of AI in software, I think people should understand that apps are going to disappear,” Pei stated unequivocally. His reasoning stems from a fundamental critique of current user experience, which he describes as “very old-school” and largely unchanged since the era of Palm Pilots. The process involves navigating lock screens, home screens, and full-screen applications, a paradigm he finds inefficient for accomplishing real-world goals.

To illustrate his point, Pei deconstructed a simple intention: grabbing coffee with a friend. Currently, fulfilling this desire requires a user to juggle multiple applications—a messaging app for coordination, a maps app for location, a rideshare app for transportation, and a calendar app for scheduling. This fragmentation creates friction. Conversely, an AI agent, understanding the user’s context and preferences, could orchestrate the entire sequence autonomously. The shift, therefore, is from a manual, app-based execution model to an intentional, agent-driven one.

From Boring Automation to Proactive Intelligence

Pei’s vision for AI integration extends beyond simple command execution, which he dismisses as “super boring.” The initial phase, already emerging in the market, involves AI performing single tasks like booking travel. However, the transformative potential lies in the subsequent stage: long-term intention learning. Here, the AI agent evolves from a reactive tool into a proactive partner. For example, if a user expresses a desire to improve their health, the device could consistently surface personalized suggestions, schedule workouts, or recommend dietary changes.

“I think it gets even more powerful when it starts surfacing suggestions for you; you don’t have to manually come up with an idea,” Pei explained. He compared this advanced functionality to features like ChatGPT’s memory, where the system builds a persistent understanding of the user. The ultimate goal is a device that “knows you very well” and acts on your behalf, anticipating needs you may not have explicitly articulated. This represents a fundamental rethinking of the human-computer relationship, moving from input-output to collaborative partnership.

Funding a Future Beyond the iPhone

This is not merely theoretical for Nothing. The company’s commitment to an AI-first device helped it close a substantial $200 million Series C funding round last year. At the time, Nothing pitched investors on a new kind of smartphone leveraging AI and personalization technology so accurate that users would not feel compelled to double-check its outputs. This focus on trust and reliability is critical, as current AI assistants often require human verification, limiting their utility. Pei’s roadmap suggests Nothing is actively developing the hardware and software infrastructure to make this agent-centric future a commercial reality.

Redesigning the Interface for AI, Not Humans

A cornerstone of Pei’s argument is the need for a new interface philosophy. Today’s smartphone interfaces are designed for human fingers and eyes—with icons, menus, and taps. For an AI agent to operate effectively, it requires an interface built for machine cognition. “The future is not the agent using a human interface. You need to create an interface for the agent to use,” Pei asserted. This means underlying systems and APIs that allow AI to interact with services and data frictionlessly, without simulating human touch gestures.

This evolution does not imply the immediate death of all apps. Pei acknowledges the current ecosystem, noting that Nothing’s own operating system even includes tools for creating mini-apps. The transition will likely be gradual. Initially, AI agents may act as a unified layer over existing apps, but the long-term trajectory points toward a native, agent-first operating system where traditional app icons are absent from the user’s view. The value shifts from the app itself to the AI’s ability to leverage the service it provides.

The Broader Industry Context and Expert Perspectives

Pei’s vision aligns with a growing industry trend. Companies like Google, Apple, and Samsung are deeply investing in on-device AI and agentic capabilities. Google’s Gemini and Apple’s evolving Siri platform aim to provide more contextual and proactive assistance. Furthermore, the rise of large language models (LLMs) has provided the technical foundation for systems that can understand complex intentions and break them down into actionable steps. Analysts suggest the next major battleground in mobile tech will be the AI agent platform, determining which company can create the most seamless, trustworthy, and useful proactive experience.

However, significant challenges remain. These include:

• Privacy and Security: An AI that “knows you very well” requires immense amounts of personal data, raising critical questions about data sovereignty and protection.
• Business Model Disruption: The app economy, driven by app stores, in-app purchases, and advertising, would need a complete overhaul.
• Technical Reliability: Building an AI that can reliably execute complex, multi-step tasks across different services without error is a monumental engineering challenge.
• User Trust: Convincing users to cede control to an autonomous agent requires unprecedented levels of system transparency and reliability.

Conclusion

Carl Pei’s prediction of disappearing smartphone apps paints a compelling picture of a more intuitive and efficient digital future. While the traditional app-centric model will persist for years, the direction of innovation is clearly toward proactive AI agents that understand context and execute tasks autonomously. This shift, championed by Nothing and other industry players, promises to redefine our relationship with technology, moving us from manual operators to collaborators with intelligent systems. The success of this vision will hinge not just on technological advancement, but on solving profound questions of privacy, trust, and economic transformation in the software landscape.

FAQs

Q1: What did Carl Pei say about the future of smartphone apps?
Carl Pei, CEO of Nothing, stated that smartphone applications are going to disappear and be replaced by AI agents that proactively handle tasks based on understanding user intentions, rather than users manually navigating through different apps.

Q2: How would an AI-first smartphone work differently?
An AI-first smartphone would focus on understanding a user’s long-term goals and immediate context. Instead of requiring the user to open specific apps, the AI agent would automatically coordinate actions across services—like messaging, mapping, and booking—to fulfill an intention, such as planning a meetup or managing health.

Q3: Is Nothing actually building such a device?
Yes, Nothing has stated that this AI-first vision was central to securing its $200 million Series C funding round. The company is actively developing hardware and software aimed at creating a smartphone where AI and personalization are the primary interface, reducing reliance on traditional apps.

Q4: What are the biggest challenges to this AI agent future?
The main challenges include ensuring user privacy and data security with highly personal AI, redesigning the app economy’s business models, achieving technical reliability for complex multi-step tasks, and building sufficient user trust to delegate important actions to an autonomous agent.

Q5: Will apps disappear immediately?
No. Carl Pei cautioned that apps will not disappear in the near-term. The transition will be gradual. The current app ecosystem will likely coexist as AI agent capabilities are developed, with agents initially acting as a layer on top of existing apps before potentially evolving into a native, agent-first operating system.