China’s Manus AI Promises Autonomous Tasks, Faces Hype and Skepticism

The invite-only AI agent claims groundbreaking capabilities but relies on existing models, raising questions about its originality and performance.

Manus AI, developed by Chinese startup Monica, is an invite-only autonomous agent designed to execute complex tasks like booking trips and creating websites with minimal user input.
The platform reportedly outperforms competitors like OpenAI's Deep Research on benchmarks but relies on existing models such as Claude 3.5 and Alibaba's Qwen, leading to doubts about its innovation.
Some early users and influencers have praised Manus for its agentic capabilities, while others have criticized its limitations, such as execution errors and reliance on pre-existing workflows.
Access to Manus is currently limited due to server capacity, with invitation codes being sold on secondary markets, fueling speculation of scarcity-driven marketing tactics.
Privacy concerns have emerged over Manus’s data handling and potential ties to Chinese authorities, highlighting the need for transparency and oversight in AI deployment.