Well, after the big old Arms race, we strongly believe that the nations of the world have a new domain to compete in - the AI race. While Open AI was doing not good but great with ChatGPT and other AI models, they faced sturdy competition from the Chinese DeepSeek AI model. And now, China being China, researchers from the country claim to have developed the world's first entirely autonomous AI agent - better known as Manus AI. The agent is already the talk of the town whenever you hear someone talking about AI models.
The launch demo of the model has garnered mixed reviews where you can see a majority of the audience praising what the Chinese have created while others finding the loopholes in the same. And yes (Always keep it in mind, not kidding), any Chinese app, AI model, whatsoever, always comes with questions related to privacy. Moving ahead, let's have a look at the working of Manus along with other necessary details.
Manus : Is it really way ahead of ChatGPT and DeepSeek?
Manus has been developed by Monica, a Chinese Startup, as per the report. And the same reports have suggested that the startup is a subsidiary of Butterfly Effect, a Singapore-registered entity. As for its working, Manus can manage complex tasks on an independent level from beginning to end with a single prompt from the user. The launch video of the model shows Manus sorting resumes, ranking candidates, and event formatting data into spreadsheets. And guess how much time it took to do so? Seconds, yes only a few seconds.
Cofounder of Manus, Yichao 'Peak' Ji said that it is not just another chatbot or workflow tool. Instead, Manus is a completely autonomous agent that plays as a medium between conception and execution. As claimed by the firm, Manus can scrape data from the web, analyze stock trends, and generate dynamic websites seamlessly. The model runs in the cloud which means that it can keep working in the background even if the user is offline.
Another stat on the official website of Manus suggests that it outperforms OpenAI's Deep Research model as per the GAIA benchmark (a tool used for comparing models). Currently, the AI model is only available on an invitation basis. And on paper, it looks pretty decent. But nothing can be said with complete certainty related to its competition with DeepSeek and Open AI models until we get to try Manus.
