Author: Zhang Feng

1. The Ups and Downs of the Manus Acquisition Case

In March 2025, a general-purpose AI intelligence product named Manus emerged suddenly. Developed by Chinese founder Xiao Hong and his team in China, this product gained global attention for its ability to "independently think, plan, and execute complex tasks end-to-end," achieving an annualized revenue quickly surpassing 125 million dollars upon launch. Within just nine months, Manus became a phenomenal star product in China's AI field and was seen as one of the representative works of China's "shortcut overtaking" in domestic AI.

However, after its stunning debut, Manus quickly initiated a fast and secretive strategic maneuver. In June 2025, the company relocated its headquarters from China to Singapore, changing its operating entity to Singapore company Butterfly Effect Pte, wholly owned by its Cayman Islands parent company. Subsequently, the Chinese team underwent significant layoffs, with only about 40 core technical personnel remaining in Singapore out of 120 employees, domestic social media accounts were cleared, and the official website blocked access from Chinese IPs. On December 30 of the same year, U.S. tech giant Meta announced it would acquire Manus' parent company, Butterfly Effect, for approximately 2 billion dollars, with founder Xiao Hong set to become Meta's Vice President as planned. This transaction makes it the third-largest acquisition in Meta's history, only behind WhatsApp's acquisition for 19 billion dollars in 2014.

After the official announcement of the acquisition by Meta, Chinese regulatory authorities quickly intervened. On January 8, 2026, Ministry of Commerce spokesperson He Yadong publicly expressed at a regular press conference that relevant departments would assess the consistency of Meta's acquisition of Manus with laws and regulations regarding export controls, technology import and export, and foreign investment. Subsequently, the National Development and Reform Commission interviewed senior executives from both parties, pointing out risks related to technology transfer and data security. Ultimately, on April 27, 2026, the Office of the Foreign Investment Security Review Mechanism (National Development and Reform Commission) legally prohibited the Meta acquisition of Manus, requiring the parties involved to revoke the acquisition transaction. This was the first AI-related foreign acquisition case publicly halted since the implementation of the "Foreign Investment Security Review Measures" in 2020, and it was the strictest review conclusion under that framework.

The core controversy of the Manus case revolves around the fact that the core technology was developed by a Chinese team within China, but through a chain of "domestically developed → offshore re-shelling → foreign acquisition," attempted to transfer control of the technology to an overseas enterprise without legal declaration for foreign investment security review. Regulatory authorities identified that the primary concern was "when, how, and what was being transferred out" — changes in registration did not exempt it from Chinese legal jurisdiction.

2. Basis of Review: Differences in Chinese and U.S. Legal Systems

China's legal system. China's foreign investment security review system is based on the "Foreign Investment Law," which establishes a management model of "pre-admission national treatment plus negative list." The "Foreign Investment Security Review Measures," released in 2020 and implemented in 2021, serves as the core execution document. According to these measures, the review covers key areas such as military, critical technologies, infrastructure, important information technology and internet products and services, and important financial services.

It is noteworthy that China's review mechanism employs a penetrating review principle. As the Manus case illustrates, regulatory authorities are not fixated on whether "the investing entity is still formally a Chinese company," but rather investigate the technology's "birthplace and growth path" — as long as the core research and development results are completed within China, even if registration changes occur abroad, it may still fall under Chinese regulatory oversight.

The U.S. legal system. The U.S. foreign investment security review system is based on Section 721 of the Defense Production Act of 1950, which formed the current framework after significantly expanding its powers through the Foreign Investment Risk Review Modernization Act (FIRRMA). FIRRMA represents the most significant expansion of foreign investment review legislation in the U.S. in 40 years, greatly enlarging the scope of review — prior to FIRRMA, the Committee on Foreign Investment in the United States (CFIUS) mainly examined transactions involving "control transfers". After FIRRMA, even non-controlling investments that do not constitute control transfers fall under the review scope.

CFIUS is an inter-agency committee led by the Department of Treasury, comprising heads from key functional departments such as Defense, State, Commerce, Homeland Security, and Justice. A key difference is that the U.S. review system institutionalizes differentiated arrangements based on "source countries" — in 2025, the Trump administration launched the "America First Investment Policy," which on one hand restricts foreign competitors' investments in critical technologies, infrastructure, and personal data, while on the other hand establishing a "fast track" pilot program for capital from allied countries.

3. Review Procedures: Proactive Declaration and Passive Tracking

China's review procedure. China's foreign investment security review process consists of three stages: the first stage is preliminary review, where a decision is made within 15 working days from the date of receipt of the compliant declaration materials on whether to initiate a security review; the second stage is general review, completed within 30 working days from the initiation of the review; if national security may be affected or is likely to be affected, a third stage special review is initiated, to be completed within 60 working days from initiation, with extensions possible if necessary. The review conclusions fall into three categories: approval, conditional approval, and prohibition of investment.

The review procedure of the Manus case is unique — it is a case of "should be declared but not declared" retrospective review. The parties to the transaction did not proactively declare, and regulatory authorities, upon noticing through public leads, legally initiated an investigation and followed the entire process before ultimately making a prohibition decision.

The U.S. review procedure. CFIUS’s review also consists of preliminary review and investigation phases. The preliminary review period lasts 45 days, and if national security risks are discovered, it enters a second investigation period of 45 days; if unresolved, it can recommend that the president intervene in the transaction, who has 15 days to make a final decision. For certain types of transactions, if the key technology produced by the U.S. target enterprise requires export licensing to the buyer's parent country, the transaction parties must submit a mandatory declaration to CFIUS.

One key difference from the Chinese review mechanism lies in the retroactive capability. CFIUS not only reviews incomplete transactions but can also conduct retrospective reviews of completed transactions — courts have even approved CFIUS's retrospective divestiture order on the HieFo-Emcore transaction, despite the acquisition having been concluded nearly two years prior. Additionally, CFIUS has a dedicated "unreported transaction team" that actively identifies transactions that may fall under its jurisdiction but have not been declared by the parties involved.

4. Review Focus: Data, Technology, and National Security

Core concerns of Chinese review. Chinese review focuses on three key points. First, critical core technology loss. The crucial evaluation in the Manus case is whether Manus’s core AI technology falls under the "information processing technology" control points in the "Catalogue of Technologies Prohibited or Restricted for Export in China." Second, data security for exiting the country. During its training and operation within the country, Manus collected a large amount of Chinese user data, raising issues about compliance with personal data exiting the country. Third, substantive control changes over formalities. Regulatory authorities penetrate Manus’s offshore structure to directly examine "when, how, and from whom the technology is transferred out."

Core concerns of U.S. review. CFIUS's core review focus is "risk-based analysis," considering the threats posed by foreign investors, vulnerabilities of U.S. target enterprises, and potential repercussions of transactions being exploited. Specifically, CFIUS pays close attention to three dimensions: first, the sensitivity of technology, assessing whether the target enterprise is involved in critical technologies covered by export control laws such as semiconductors, AI, and quantum computing; second, data risks, examining whether the involved transaction would lead to a significant flow of sensitive personal data into foreign adversaries' hands; third, differential treatment for source countries, applying stricter differentiated review standards for investments from countries like China. In addition, CFIUS review is increasingly intertwined and aligned with U.S. domestic industrial policy — the core controversies surrounding the acquisition of U.S. Steel by Nippon Steel have extended beyond traditional national security realms to include labor rights and industrial competition.

5. Penalty Measures: Prohibition and Divestiture

China's penalty framework. According to the "Foreign Investment Security Review Measures," prohibiting investment is the strictest review conclusion — the measures explicitly state that "investment must not be implemented if prohibited." If the parties refuse to comply, the mechanism office can order a time limit on the disposal of shares or assets and include their negative credit records in the national credit information system, implementing joint sanctions.

The execution requirements for the Manus case are very specific. At the equity level, if Meta has completed the equity transfer, it must return all Manus shares held to the original shareholders or domestic entities, completing business and offshore entity registration changes. At the financial level, Meta must fully return the approximately 2 billion dollars already paid, with the foreign exchange management department monitoring the capital path to prevent capital outflow under the pretext of terminating the transaction. At the data and technology level, Meta must delete all acquired domestic data, Manus must restore data local storage, terminate all technology authorizations and code transfers to Meta, and all technical personnel stationed by Meta must withdraw. The mechanism office will gather relevant departments like development and reform, commerce, cyberspace, and foreign exchange for on-site verification.

The U.S. penalty framework. CFIUS's penalty measures have also shown an escalation trend. In recent years, CFIUS's enforcement has reached new records, with total fines nearing 88 million dollars, with individual fines reaching up to 60 million dollars. Before 2024, CFIUS had only publicly disclosed six penalties, with the largest being 1 million dollars. In February 2026, the U.S. Department of Justice filed the first enforcement lawsuit based on the Defense Production Act regarding a CFIUS divestiture order. In terms of retroactive effect, the U.S. is even more aggressive than China — in January 2026, President Trump issued an executive order requiring Chinese-controlled HieFo to divest its digital chip business acquired from U.S. Emcore within 180 days, despite the acquisition having been completed nearly two years before.

6. Focus Areas: Commonalities and Differences

Common focus areas. The core intersection of the review mechanisms in China and the U.S. is concentrated in the precision technology sector. The semiconductor industry chain is the area with the highest frequency of examination and the highest rate of rejection on both sides — prior to halting the Manus acquisition, CFIUS had already formed a de facto comprehensive blockade on semiconductor transactions through various cases. AI and related intelligent technologies are a common top priority — the Manus case marks China's first prohibition of investment decision in the AI sector, contrasting with the U.S.'s longstanding strict review in this area.

Differentiated focal points. China's review emphasis is more on "technology flowing out of China" rather than "foreign investment entering China." The regulatory pressure in China primarily focuses on preventing the "domestically developed, offshore cashing out" model — the loss of core technology assets has become the foremost concern in reviews. In contrast, U.S. reviews focus more on limiting foreign capital from acquiring U.S. technology, having integrated geopolitical variables into the evaluation framework earlier than China, and clearly incorporating ally relationships into procedural innovations — the "fast track" processes accelerate review timelines for investors from specific countries.

7. Global Impact: Regulatory Reflection of the Tech Race

The Manus case coincided with the launch of DeepSeek V4, creating a meaningful contrast. The rise of DeepSeek V4 signifies that domestic AI is capable of standing firm in the highest level of the global sequence; while the halt of the Manus acquisition means the state will not allow the loss of core AI technology assets through mergers. Together, they outline the dual aspects of China's AI strategy: one being full support for independent innovation to grow and strengthen, and the other being maintaining the bottom line for core asset security.

The global impact of the Manus case manifests in at least three layers. First, the technological rift between China and the U.S. has deepened, forcing multinational capital to reassess cross-border merger risks in the AI sector. Second, global foreign investment regulation shows a trend of collaborative expansion. From the U.S. to the EU, from the UK to Japan, major economies are synchronously tightening foreign investment review frameworks in critical fields. Third, security reviews are replacing traditional antitrust reviews as the primary regulatory barrier facing frontier technology merger transactions.

8. Future Development: Trends and Corporate Responses

Regulatory trends. Looking ahead, it is expected that both China and the U.S. will continue to strengthen regulatory efforts in the field of foreign investment security review. In China, the Manus case, as the first foreign acquisition case in the AI field to be halted, will inevitably promote further refinement of the review concerning critical technology and data security. In the U.S., CFIUS will continue to develop along the lines of differentiated treatment for investment sources, strict control of sensitive areas, and strengthening industrial chain associations. The national security law on foreign investment passed by Congress in December 2025 further expands the toolbox of legal instruments for technology protection in the U.S.

Corporate compliance recommendations. For enterprises engaged in frontier technology cross-border transactions, the Manus case provides clear compliance warnings.

— Proactive pre-approval declaration. Chinese regulatory authorities hold a zero-tolerance attitude towards "declared but not reported," and proactive declaration is the most effective way to reduce regulatory uncertainty. Transaction parties should negotiate with the security review mechanism office to manage regulatory risks through transparent compliance paths.

— Cautious design of transaction structures. Treat compliance review as a red line in transaction structure design. Adopt phased transaction arrangements, making regulatory approval a prerequisite for closing. Any attempt to evade review through offshore structures may be negated by penetrating review.

— Technology export compliance management. In cross-border transfers involving critical technology, compliance assessments should be made against regulations such as the "Catalogue of Technologies Prohibited or Restricted for Export in China" in advance, applying for technology export licenses when necessary.

— Data cross-border compliance arrangements. Technical transfers involving the processing and exit of Chinese user data must fulfill statutory procedures such as data cross-border security assessments.

— Special considerations for investments in the U.S. In addition to focusing on CFIUS’s own review, investments in the U.S. must also cautiously assess whether the export control systems of both sides may lead to dual compliance conflicts, and proactively design risk allocation mechanisms in transaction documents.

The comparison between China and the U.S. reveals that the halt of the Manus acquisition is not an isolated event but a typical case under the trend of tightening foreign investment reviews across major global economies in the frontier technology sector. The review systems of the two countries have different focuses and measures, yet their underlying logic is highly consistent: core technology is essentially a strategic national asset, and the sovereignty of core data cannot be ceded. Understanding this basic logic may be more important than being familiar with the legal text of both countries.

免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。