- Manus AI is a fully autonomous intelligent agent developed in China.
- It outperforms models such as GPT-4 and OpenAI's Deep Research in several aspects.
- It is in invitation-only beta, which increases its exclusivity.
- Its development has raised concerns about privacy and transparency.
China is once again taking center stage in the race to lead artificial intelligence. After the success of DeepSeek, a A new platform called Manus AI is challenging giants like OpenAI and Google DeepMind.. The key? Your ability to act completely autonomously, taking human-machine interaction to another level.
Numerous technological media and experts already confirm it.: Manus AI could be one of the most significant advances in the evolution towards general artificial intelligence.Despite the doubts and criticism, its launch has generated unprecedented excitement, especially in Asia, where it has quickly gone viral.
What is Manus AI and who is behind its development?
Developed by Chinese startup Butterfly Effect, Manus AI is an artificial intelligence platform that promises to become the world's first truly autonomous AI agent. Inspired by the Latin word "manus" (hand), it symbolizes your ability to turn thoughts into tangible actions.
Its operation is based on advanced linguistic models as Claude (from Anthropic) and Qwen (from Alibaba), although its architecture goes further. It uses a multi-agent system where various specialized models cooperate to perform complex tasks from start to finish without human intervention. This adds to the innovations that have emerged in Chinese AI that predicts diseases through blood tests.
According to Ji Yichao, the company's chief scientist, Manus is not just another chatbot with programmed commands, but an AI capable of analyzing and executing entire processes autonomously, from everyday actions such as shopping, to more complex ones such as programming video games, planning trips, or investing in the stock market.
What Manus AI can do in practice
Manus AI's capabilities have been showcased in multiple demonstrations. In one, for example, analyzed a compressed package with dozens of resumes, extracted useful information such as skills and experience, compared it with current employment trends, and then proposed selection strategies based on different criteria.
In another exercise, when receiving an ambiguous request such as “find me an apartment in San Francisco”, AI processed information on security, rental prices, weather, and quality of life in different areas to generate personalized recommendations., something that could revolutionize the way users navigate similar applications.
This is possible thanks to its asynchronous approach in the cloud, which allows you to perform continuous operations in the background. The user only receives a notification when the process or investigation is complete, avoiding the need for constant interaction.
In addition, Manus IA optimizes human-machine collaborative work by dividing tasks into parts that are handled by specialized models. This method reduces errors and speeds up workflows compared to traditional AIs, which typically operate in a monolithic manner.
Advantages over other artificial intelligence models
According to several sources such as Wired, Euronews, and Hipertextual, Manus AI has outperformed OpenAI's Deep Research in benchmarking tests such as GAIA, a specific evaluation to measure the autonomy and performance of general-purpose assistants.
These tests indicate that Manus not only matches, but surpasses AI capabilities such as GPT-4 and Gemini.While there is controversy over the transparency of the methods used in these evaluations, the truth is that the technology community is paying attention.
A big difference from other emerging AI is that Manus is not exclusively aimed at conversational tasksIts approach is based on autonomous actions without the need for constant feedback, making it a much more robust option for work and business contexts. This could make a significant difference in how Asian super apps are used in the future.
Current limitations: restricted access and technical issues
Although the promises are great, Manus AI is still in beta and is available by invitation only.This has fueled interest to the point where access codes are being traded for astronomical amounts in online auctions, even reaching 12.000 euros on Chinese platforms.
Despite the enthusiasm, major failures have also been reported. Experts such as Pierre-Carl Langlais have pointed out that AI can get stuck in infinite loops. or failing on long tasks composed of multiple steps, as occurred in a 20-step workflow where Manus failed at step 18 after nearly an hour. This highlights that, while it has impressive potential, significant performance improvements are still needed.
Likewise, other users have noticed that The platform sometimes omits easily accessible information or responds with errorsThese issues reflect that, while the Manus's range is impressive, its accuracy and reliability still need further improvement.
Privacy and transparency concerns
One of the most sensitive issues surrounding Manus AI is its origin: China. According to the country's national intelligence law, All technology companies are required to cooperate with state intelligence agencies., which raises suspicions about the possible misuse of the collected data.
Bradford Levy, a professor at the University of Chicago, warns that The lack of transparency about how Manus works internally and how it manages data should be cause for concern. for any international user interested in using it.
In fact, some voices in the academic field point out that Manus could become a covert tool for gathering sensitive information globally., taking advantage of their attractive design and technological promise to allow users to freely share personal or business information. This phenomenon may be similar to the scandal involving Seagate hard drives used in certain workplaces.
Future outlook: What impact could it have on Europe?
According to data cited by Marketplace Innovation Offices and Amazon Web Services, 44% of Spanish companies already use AI in their operationsFurthermore, 81% of them believe that generative AI will be a strategic priority by 2025.
If Manus can expand beyond China and offer a cost- and performance-competitive model, Its adoption in Europe could grow rapidly, especially in areas such as data management, legal advice, and business process automation. However, this will depend on the platform's ability to adapt and overcome current limitations.
However, for that to happen Manus must improve its transparency, technical reliability and ensure ethical use of data.Only then will you be able to truly compete in a technological environment where trust is as important as innovation.
Todo indica que The battle between powers to dominate AI is far from over., and with projects like Manus, China continues to demonstrate its ability to lead the next generation of artificial intelligence. While access is limited and its capabilities are still being evaluated, its technical principles represent a clear evolution compared to current systems. If operational stability and legal guarantees are achieved, we could be witnessing one of the most transformative tools of the near future.
I am a technology enthusiast who has turned his "geek" interests into a profession. I have spent more than 10 years of my life using cutting-edge technology and tinkering with all kinds of programs out of pure curiosity. Now I have specialized in computer technology and video games. This is because for more than 5 years I have been writing for various websites on technology and video games, creating articles that seek to give you the information you need in a language that is understandable to everyone.
If you have any questions, my knowledge ranges from everything related to the Windows operating system as well as Android for mobile phones. And my commitment is to you, I am always willing to spend a few minutes and help you resolve any questions you may have in this internet world.