Amazon co-founder MacKenzie Scott has donated over $19 billion to charity in just five years
Diamond batteries powered by nuclear waste promise 28,000 years of clean energy
More Than Just a Chatbot
Unlike typical AI systems, Manus operates with a higher degree of autonomy: it goes beyond merely answering queries or drafting summaries. It independently plans, executes, and completes complex tasks without human input.
Developed by the low-profile firm Manus AI, this new technology represents a significant step forward in the realm of general-purpose autonomous agents. It operates without needing explicit instructions, effectively thinking on behalf of the user…
The operation of Manus is structured into three phases, starting with planning (crafting a detailed action plan based on the given goal), followed by autonomous execution (it leverages its internal resources to carry out tasks whether it involves researching, analyzing, or creating content), and finally completion without supervision. It’s important to note that no human validation is required between these stages.
Upon its debut, Manus demonstrated that it’s more than just marketing hype. Public demonstrations showcased its prowess in several areas like advanced financial analysis. For instance, Manus produced a comprehensive stock report including a SWOT analysis (strengths, weaknesses, opportunities, threats), a competitive evaluation, and investment recommendations, all without human intervention.
It can also create entire websites: the AI gathered data on a public figure, wrote their biography, crafted a website in HTML/CSS, and even deployed it online while managing any technical glitches on its own.
Impressive Demonstrations
On its website, there’s a benchmark with OpenAI that looks quite promising, though it’s hard to verify independently. Although the model is open source, not everyone can test it freely; registration is required to receive an access code.
What really sets Manus apart is its ability to think like a project manager. If it’s as real as promised, this could be a game-changer, positioning China at the forefront of the autonomous agent race, while companies like OpenAI and Anthropic are still refining their intelligent agent models. Given the capabilities highlighted, it’s no wonder the USA is concerned, already considering restrictions similar to those applied to DeepSeek!