Summary of Key Points
Manus is a controversial AI startup: it did not develop its own large-scale models but instead created intelligent agent products by integrating overseas APIs and open-source frameworks. Despite this, it generated over 100 million in revenue within just 8 months. In less than four years, the company has experienced three significant turns of fate—starting from a residential building in Wuhan, moving to the negotiation tables of Meta, being halted by regulatory authorities due to an acquisition attempt, and finally being acquired by Tencent. Its sudden success was driven by its ability to meet users' needs for AI-powered productivity, but it has always been surrounded by controversy over its business model. Its rapid capitalization led to a soaring valuation, yet its lack of technical barriers makes it difficult for the company to sustain itself independently. Manus's story serves as both a successful example of commercializing AI applications and a reminder of the limitations of a "lightweight" business approach.
Analysis in Detail
1. The Root of the Controversy: A "Shell Company" or a "Productization Expert"?
The core issue surrounding Manus is whether it possesses genuine technical barriers:
- Critics argue it's a shell company: It lacks its own underlying large-scale models and relies on Anthropic’s Claude API, along with open-source frameworks, to create automated functions. They point out that since similar products can be replicated by the open-source community within hours, the technical barrier is relatively low.
- Supporters see it as a productization winner: Ordinary users don’t care which model is used; they only want solutions to their problems (e.g., quick resume analysis or website generation). Manus encapsulates complex technology into user-friendly tools, similar to how Apple can sell smartphones without manufacturing chips or Tesla can produce electric cars. In this case, the product experience and engineering capabilities themselves constitute a competitive barrier.
This debate reflects a fundamental split in the AI industry: should one focus on developing underlying models first or on addressing user needs immediately? Manus chose the latter, quickly turning AI capabilities into practical products for the general public.
2. How Did It Make Quick Money? By Seizing Three Critical Opportunities
While the AI industry is generally loss-making, Manus managed to generate over 100 million in revenue within 8 months by hitting three key points:
- Addressing User Pain Points: While ChatGPT is primarily for chatting, Manus can perform practical tasks (e.g., automatic stock analysis or website creation), meeting the needs of businesses and premium users for AI-powered productivity.
- Targeting the Right Market: It focused on overseas users, who are more willing to pay for valuable tools. For instance, its invitation codes were highly sought after, indicating demand from international customers.
- Effective Marketing: It created scarcity through invitation codes and leveraged social media buzz and controversial topics to make the concept of a "universal intelligent agent" a hot topic among investors, thus attracting attention despite technical concerns.
However, its profit model has risks: with daily computing costs of $500,000 and monthly revenues of 8.33 million, it incurs a significant cash flow deficit, forcing the company to rely on sales to survive.
3. A Lightning-fast Capital Operation: From a Valuation of 500 Million to a Sale for 2 Billion in 8 Months
Manus’s team excels at capital and geopolitical arbitrage:
- Rapid Financing: Three months after its success, it secured $75 million from top Silicon Valley venture capital firm Benchmark, with a valuation of 500 million—investors recognized the value of selling products over developing models.
- Geopolitical Shift: To appeal to American investors, it moved its headquarters to Singapore within two months and laid off Chinese employees, cutting ties with the Chinese market and gaining growth overseas.
- Quick Sale to Meta: Zuckerberg saw the potential of Manus’s AI capabilities to complement Meta’s ecosystem, and the acquisition was completed in just 10 days for $2 billion. The team quickly realized the highest possible valuation before selling the company.
This aggressive capital strategy allowed Manus to find a buyer before the bubble burst, but it also set the stage for future regulatory challenges.
4. Regulatory Halt: Why Was the Meta Acquisition Blocked?
In April 2026, the National Development and Reform Commission halted Meta’s acquisition of Manus on grounds of national security concerns, with the decision based on where the actual research and development took place, not just where the company was registered. Although Manus moved its headquarters to Singapore, its core R&D team remained in China, indicating that key activities were still conducted there. Given the sensitive nature of AI and data security, foreign acquisitions could lead to the loss of core capabilities, leading to the regulatory intervention.
This ban put Manus in a difficult position: it couldn’t return to the U.S., had lost access to the Chinese market, and its Singapore-based headquarters lacked essential R&D resources. It was only when Tencent acquired the company that it found a new lifeline.
5. Why Did Tencent Acquire It?
Tencent’s decision to buy Manus for $2 billion was driven by several factors:
- Traffic Potential: Manus’ revenue surged from 100 million to 400–500 million in the second half of 2026, demonstrating the viability of its products. Tencent’s platforms (WeChat, QQ) offer vast user bases that could boost Manus’s revenue significantly.
- Completing the AI Ecosystem: Tencent’s own large-scale models (e.g., Hunyuan) lag behind those of ByteDance and Baidu, and it needed a strong application-layer product. Manus’s intelligent agents provided an ideal addition to its ecosystem.
- Early Investment Return: Tencent had invested in Manus when its valuation was only 85 million, earning a 23-fold return on that investment. Acquiring Manus allowed Tencent to further expand its AI capabilities and strengthen its ecosystem.
Lessons for the AI Industry
Manus’s story highlights two possibilities within the AI industry:
- Positive Aspect: The application layer can quickly establish a viable business model. Companies don’t necessarily need to focus on developing their own large-scale models; by packaging existing technologies into useful products, they can still make money, offering a "lightweight" approach for smaller AI startups.
- Negative Aspect: Models without substantial technical barriers are vulnerable to competition. Manus’s approach is susceptible to substitution if larger companies (like OpenAI) enter the market with their own AI solutions. In the long run, independent control over underlying models will be crucial for sustainable success.
In summary, Manus demonstrated that the application layer can achieve rapid success, but it also underscores that technical prowess is essential for enduring success in the AI industry. While a compelling story can drive short-term valuations, it is technology that ensures long-term competitiveness.