In just a few days, Manus' reputation was like riding a roller coaster, soaring to the sky at one point and being 'knocked down from the mortal world' at another. On March 6th, Manus, known as the "world's first universal AI intelligent agent," was released, sparking a heated discussion both domestically and internationally overnight. Words of praise such as "the next DeepSeek" and "the GPT moment of AI intelligent agents" flooded in, and the entire network was eagerly seeking an invitation code at a high price. Shortly after leaving the circle, doubts followed. After experiencing it, some people expressed that it did not meet the official demonstration effect and external expectations, and was not enough to be called a "universal AI intelligent agent". Some even suspected that Manus was just a "shell" product. Moreover, there is not much discussion about this product overseas, and it is only a matter of "self promotion" domestically... Within just two or three days of Manus' release, there has been such a serious polarization in external evaluations, which is directly related to product performance and user experience. In this wave of public opinion, it is unclear whether the developers of Manus deliberately inflated their value through "hunger marketing", and whether they can "rapidly evolve" the product in a short period of time to truly establish a foothold remains to be seen. But regardless of the quality of Manus products, it is not a bad thing that there are constant voices of admiration and skepticism. On the one hand, it shows that Chinese netizens are eager to see Manus replicate the success of DeepSeek and take the lead in the application of global general AI agents, thus proving the super strength of China's AI industry; On the other hand, it also shows that the majority of countrymen have a clearer understanding of AI applications. Many people give objective evaluation only after participating in the test. They do not believe in the popularity of the Internet. They firmly believe that AI agents are "mules or horses for walking", demonstrating a pragmatic attitude. From this perspective, whether Manus has "overturned" is not important or scary, but rather indicates to some extent that with the rapid development of the domestic AI industry, the AI literacy and exploratory spirit of the general public are widely popularized. It should be said that at the beginning of 2025, the global impact brought by DeepSeek shows that the domestic AI industry is transforming from technological catch-up to local leadership. It has reached the international advanced level in low-cost training and long thinking reasoning, and various fields such as big languages and multimodal big models are also flourishing. What is even more gratifying is that there is a strong innovation atmosphere in the domestic AI industry of "chasing after each other and seizing the day", which has formed confidence and consensus in daring to open source. Various new products continue to challenge the global AI performance testing list. Various teams treat DeepSeek as both a teacher and a competitor. In addition to joy, questioning and staying calm about AI applications is also a good thing. Because we must see that the global AI competition has entered a white hot stage. The United States continues to lead in basic theory, high-end chips, and other fields. With the triple support of technological innovation, capital investment, and policy system, product iteration continues to accelerate. And other developed countries are also accelerating their layout, competing for technological and industrial application opportunities. In this competition that concerns the future, although China has an advantage in the application level, problems such as shortcomings in basic research, dependence on high-end devices, and a shortage of top talents are still prominent. It is not easy for truly disruptive and revolutionary technological breakthroughs and application innovations to occur frequently in a short period of time, following DeepSeek, as the saying goes, 'Without taking small steps, one cannot reach a thousand miles.'. Standing at a critical moment in the era of artificial intelligence, the domestic AI industry not only needs the courage to "overtake on the bend", but also the determination to "sharpen the sword for ten years". From breakthroughs in basic algorithms to deepening scene applications, from building computing power networks to cultivating talent teams, every link requires a deep commitment to continuous investment. Only by taking one step at a time can we take the initiative in the global AI competition and truly achieve a new leap from being a "follower" to a "leader". (New Society)