Meta Expands AI Push With Manus Desktop App Amid Agent Boom

Artificial intelligence startup Manus, recently acquired by Meta, has introduced a new desktop application designed to bring its AI agent directly onto users’ laptops. The launch marks a significant shift from its earlier cloud-only model, enabling deeper integration with personal devices and local workflows.

Previously, Manus’s general-purpose AI agent operated primarily through a web interface, handling tasks in the cloud. With the new desktop version, a feature called “My Computer” allows the agent to interact directly with files, applications and tools stored on a user’s machine. This enables the AI to execute more complex, multi-step tasks without relying entirely on remote infrastructure.

The move places Manus more directly in competition with OpenClaw, a rapidly growing open-source AI agent that runs locally on user devices. OpenClaw has gained widespread attention in recent months, with Jensen Huang describing it as potentially “the next ChatGPT,” highlighting the growing excitement around autonomous AI systems.

Unlike OpenClaw, which is free and open-source, Manus operates on a subscription-based model. However, its enterprise-grade positioning and integration capabilities may appeal to users seeking more structured and supported AI tools. The platform already connects with services such as Google Calendar and Gmail, and now extends its reach into local computing environments.

According to the company, the “My Computer” feature allows the AI agent to read, analyze and modify files, as well as launch and control applications. Users can assign tasks such as organizing large datasets, managing thousands of images, or even building software applications through integrated coding tools. These capabilities represent a shift toward AI systems that not only assist but actively execute workflows.

The growing adoption of local AI agents reflects a broader trend in the industry, where companies are moving beyond traditional chat-based interfaces toward more autonomous systems capable of planning and acting independently. This evolution is reshaping how users interact with software, blurring the line between tools and digital collaborators.

However, the expansion of AI agents into personal devices has also raised concerns among experts. Granting AI systems access to local files and applications introduces potential risks related to data privacy, security breaches and unintended actions. In response, Manus said its system includes safeguards requiring explicit user approval before executing tasks, offering options such as one-time permission or ongoing authorization for trusted actions.

Meta’s investment in Manus is part of a larger strategy to strengthen its position in the rapidly evolving AI landscape. The company announced its acquisition of Manus in late 2025, aiming to integrate the startup’s autonomous agent technology into its broader ecosystem, including its AI assistant offerings.

The deal has also attracted regulatory attention. Manus, originally founded in China before relocating its headquarters to Singapore, is reportedly under review by Chinese authorities for potential compliance issues related to technology controls. Meta has stated that the acquisition complies with applicable laws and expects the matter to be resolved.

As competition intensifies among major technology players such as OpenAI, Meta, and others, the rise of AI agents capable of operating across both cloud and local environments signals a new phase in the evolution of artificial intelligence—one where software increasingly acts on behalf of users rather than simply responding to them.