Claude agent on your phone: tasks now run in the cloud

Anthropic has updated its Claude Cowork agent: tasks now keep running in the cloud even after the laptop is closed. You can start a task and follow it from your smartphone, while the agent works in the background and sends the result when it finishes. The update is part of Anthropic's broader pivot toward mobile AI agents.

AI-processed from Wired; edited by Hamidun News
Claude agent on your phone: tasks now run in the cloud
Source: Wired. Collage: Hamidun News.
Listen to article

Claude now works on smartphones: in July 2026, Anthropic updated its Claude Cowork agent so that tasks keep running in the cloud even after the laptop is closed.

What changed for users

Before this update, agentic tools, including Claude Cowork, required an active session. Close the browser or put the laptop to sleep, and the task was cut off. That was tolerable for short requests, but for long, multi-step work such as research, data analysis and workflow automation, being tied to an open device was a serious limitation.

Anthropic has moved the computation to the server side. Users can now start a task on a laptop, leave the house, and a few hours later receive the finished result on their smartphone, without having to keep the computer on.

The mobile interface lets users launch agent tasks directly from a smartphone, track progress in the background, receive completion notifications and, if needed, give the agent extra instructions while it works.

Why Anthropic is betting on the smartphone

The mobile extension of Claude Cowork is part of Anthropic's larger strategic bet on agents as its core product. The company has consistently pushed Claude toward autonomous task execution rather than simply generating answers.

The smartphone changes how people interact with an agent. Previously the agent was a workstation tool: you started it from a computer and monitored it from a computer. Now it becomes a persistent background assistant: start it from your phone, go to a meeting, come back, and it's done.

Competitors are heading the same way by different routes: OpenAI is deepening agent capabilities in mobile ChatGPT, Google is building Gemini into Android at the OS level, and Apple is integrating AI features into native iOS apps. Anthropic is choosing the niche of cross-platform continuity: the task lives in the cloud, not on the device, and the user connects to it from anywhere.

What is Claude Cowork?

Claude Cowork is Anthropic's agent platform for autonomously carrying out multi-step tasks: researching a topic, writing and running code, working with files, and interacting with external services. Unlike a regular chat mode, where the user guides each step, Cowork receives a goal and plans the actions to reach it on its own.

Before the update, the agent's autonomy was limited by the length of the active session. The practical ceiling was a few minutes or tens of minutes, until the device went to sleep. Background execution removes that ceiling.

What this means

Anthropic is turning Claude from a chatbot into an agent that works continuously, like an app you don't need to keep open. If the "always-on agent" pattern becomes an industry standard, 2026 could be a turning point: the AI assistant moves beyond the desktop and becomes part of everyday mobile life.

Did Claude previously require an open laptop?

Yes. Before the July 2026 update, agentic tools, including Claude Cowork, required an active session. If you closed the browser or the device went to sleep, the task was cut off.

Can the Claude agent run on a smartphone?

Yes. In July 2026 the Claude Cowork interface was brought to smartphones, and tasks now run in the cloud regardless of whether the app is open.

What was the Claude agent like before?

Before the update, Claude Cowork required an active session. If you closed the browser or the device went to sleep, tasks were interrupted.

Story

Key Questions Answered

What is Claude Cowork?

Claude Cowork is Anthropic's agent platform for autonomously carrying out multi-step tasks: researching a topic, writing and running code, working with files, and interacting with external services. Unlike a regular chat mode, where the user guides each step, Cowork receives a goal and plans the actions to reach it on its own. Before the update, the agent's autonomy was limited by the length of the active session. The practical ceiling was a few minutes or tens of minutes, until the device went to sleep. Background execution removes that ceiling.

Did Claude previously require an open laptop?

Yes. Before the July 2026 update, agentic tools, including Claude Cowork, required an active session. If you closed the browser or the device went to sleep, the task was cut off.

Can the Claude agent run on a smartphone?

Yes. In July 2026 the Claude Cowork interface was brought to smartphones, and tasks now run in the cloud regardless of whether the app is open.

What was the Claude agent like before?

Before the update, Claude Cowork required an active session. If you closed the browser or the device went to sleep, tasks were interrupted.

ZK
Zhemal Khamidun — CPO at Alpina Digital, product manager of AlpinaGPT. Building AI products since 2019. Hamidun News is a pet project that grew into a production aggregator with 6,000+ readers.
✦ Unified AI Workspace · Nomad

Test Frontier LLMs & Autonomous Agents in One Workspace

Run Claude Opus 5.5, GPT-6, Gemini 3.1, and DeepSeek side-by-side in Nomad AI (nomadnet.ai) — unified model routing, deep research, and multimodal workflows under one subscription.

Try Nomad AI →Launch Nomad Agent
What do you think?
Loading comments…