Manus AI. More Than Just a Chatbot or LLM! AI News – #2

Manus AI is a new autonomous artificial intelligence agent developed by the Chinese startup Monica, which debuted on March 6, 2025. The name Manus comes from the Latin word for “hand,” symbolizing its ability to transform users’ thoughts into specific actions.

Manus is a newcomer to the Chinese artificial intelligence scene: it’s an advanced AI agent that transforms user thoughts into actions. Unlike traditional chatbots, which primarily process text and generate responses, Manus is designed to autonomously perform tasks without constant human supervision.

Manus, created by the Chinese startup Monica, was officially launched on March 6, 2025, and quickly attracted the attention of the tech industry. Its ability to autonomously plan, make decisions, and execute tasks represents a significant step towards achieving artificial general intelligence (AGI). The introduction of Manus caused a stir in the market, leading to situations as bizarre as the emergence of a black market for system access codes, with prices reaching as high as $14,000!

Manus AI operates autonomously, utilizing advanced machine learning algorithms and language models, such as Claude 3.5 Sonnet and Qwen, to accomplish complex tasks with minimal human intervention. Its architecture is based on a multi-agent system, which allows for iterative task processing in a manner similar to human reasoning. During this process, it analyzes commands, selects appropriate tools, executes commands in an isolated environment, and ultimately delivers results in an organized format.

Whereas standard LLMs, such as ChatGPT, respond to queries and conduct conversations based on provided information, Manus AI possesses the ability to independently initiate actions, analyze data, and dynamically adjust its strategies in real-time. Manus is designed to autonomously make decisions and execute complex tasks without continuous user interaction. It can, for example, independently plan trips or create animations, functioning as a personal assistant. Manus goes beyond standard chatbots by not just responding to commands but taking action on its own. Unlike traditional chatbots that wait for specific instructions, Manus proactively analyzes user needs and initiates tasks, seamlessly integrating with various systems and tools to get things done.

Here’s how Manus’s creators describe the solution:

We’ve already mentioned that currently available LLMs require direct instructions from users. Manus AI is designed to independently initiate tasks, evaluate new information, and dynamically adjust its approach. It acts as a manager, overseeing multiple specialized components (sub-agents), which allows it to execute complex, multi-stage tasks without human oversight. It can work asynchronously, performing tasks in the background and notifying users of the results only upon completion.

Manus AI is capable of performing tasks in various areas; its creators cite examples such as financial analysis, employee recruitment, and website creation. The agent can autonomously analyze resumes, select employees based on market trends, taking into account analyses and accessed reports. Similarly, in the case of coding, it can independently create a website, prepare content for it, and implement necessary fixes in case of any hosting issues.

Sounds interesting? Check out the example use cases of Manus AI published by the creators on their website.

Currently, Manus AI is in a closed beta phase, which means that access is limited and requires a special invitation from test participants. The creators explain the limited availability by the need to conduct system stress tests and identify potential issues before wider release of the tool. Those interested must wait for the opportunity to test Manus; for the time being we’ve encountered both excitement and skeptical comments in the community.

Despite its impressive capabilities, Manus AI has faced some challenges. In tests by Business Insider, it was found to generate false data and copy content. For example, during a sentiment analysis on federal workforce cuts, it created simulated data and fake social media reactions without user consent. All this made the results completely unreliable.

People are also raising concerns about data privacy and security. Manus was developed in China, where local laws require companies to share data with government agencies. Experts warn that users should think twice before sharing sensitive information on platforms like Manus, as there’s a risk that their data could be accessed by various institutions.

Manus AI is a big leap forward in autonomous AI, with the potential to change the way we interact with technology. But it’s not quite there yet – issues with accuracy, privacy, and limited availability mean there’s still a lot of work to be done before it becomes a truly reliable and accessible tool.

Interested in cutting-edge AI solutions? Drop us a message at [email protected] and subscribe to our newsletter!