The Death of Icons: Why 80% of Mobile Apps Will Soon No Longer Be Needed

# The Death of Icons: Why 80% of Mobile Apps Will Soon Become Unnecessary

The cardboard box of applications on the smartphone screen won't last long. This is suggested by OpenClaw founder Peter Steinberger, who recently presented investors at Y Combinator with a provocative thesis: the operating system of the future won't need icons—it will need intentions. And he is not exaggerating. According to his calculations, eighty percent of existing applications will simply stop being opened manually, because artificial agents understanding what we need will open them for us before we even realize it ourselves.

This sounds like science fiction, but the mechanisms of this transformation are already working. The industry is moving simultaneously in three directions, and each one is remaking what we used to call a mobile interface. The first is restructuring at the logic level.

OpenClaw doesn't rely on a single neural network answering questions. Instead, the project has built a coordination system of many specialized agents working in the background. Think of it as a swarm of bees, where each individual is responsible for its own task: one agent checks email, another books tickets, a third transfers money.

A distributed control architecture exists above the level of individual applications—it doesn't call them, it orchestrates them. Yes, while the success rate of multi-step operations drops in more complex scenarios, the principle is already proven: a fundamentally new level of execution can be built above applications.

The second path is more crude, but effective. ByteDance, the Chinese owner of TikTok, chose a strategy of visual capture. Instead of waiting for developers to open their applications' APIs, the company simply taught AI to look at the screen and independently "use" applications like a human.

The bot presses buttons, scrolls, enters text—all through visual interface recognition. ByteDance even launched its own smartphone with system AI for this. The idea revealed its potential: the boundary between human control and automation is blurred.

But reality is harsher than the dream. With the current power of mobile chipsets (about 30 trillion operations per second), delays when switching between applications reach three seconds, and the reliability of operations barely exceeds the fifty percent threshold. However, the vector is set: hardware limitations are temporary, architectural solutions are eternal.

The third, deepest shift occurs at the operating system level. Here, fundamental economic relationships between technology giants are undergoing transformation. For ten years, Google paid Apple approximately two hundred billion dollars annually so that the default search bar would show its results. Access to the digital environment was controlled by search. Now, as reported by Bloomberg, the situation is reversing: Apple may start paying Google for Gemini integration—an artificial intelligence for system smartphone management. This is not just a change of partner, it is a redefinition of what "access" means. If access used to be a search bar, then tomorrow access is a conversation with intelligence built into the operating system itself.

For now, applications won't disappear anywhere. But their role is changing catastrophically. From interaction points that the user consciously opens, they are becoming "digital pipelines" through which information called up by AI flows. You won't open navigation because the agent will have already called up the necessary coordinates; you won't search for flights because the system already knows you need to get to the airport tomorrow at eight in the morning. An icon on the screen is an artifact of the past. The future doesn't need icons because it speaks.