China's Manus AI Claims to Be the First Fully Autonomous AI Agent

The newly launched platform promises human-level task execution but faces skepticism over performance and reliability.

Manus AI, developed by Beijing-based startup Monica AI, is being touted as the first fully autonomous AI agent capable of executing complex tasks without human oversight.
The platform integrates existing large language models like Anthropic’s Claude and Alibaba’s Qwen, fine-tuning them for autonomy rather than building a proprietary foundational model.
Key innovations include a multi-agent architecture, asynchronous cloud-based operation, advanced tool usage, and contextual memory to streamline task execution.
Early user feedback highlights impressive capabilities but notes issues such as slow performance, task errors, and incomplete executions during its beta phase.
Competitors like OpenAI’s Operator and Anthropic’s Claude are seen as more polished but less ambitious, while future advancements from Google and xAI could intensify competition.