Public reporting on Meta’s acquisition of Manus is not merely a transaction-specific development. It is a reminder that cross-border AI transactions are increasingly assessed by reference to substance, strategic sensitivity and regulatory perception, rather than structure alone.
According to public reporting, Chinese authorities blocked or ordered the unwinding of Meta’s acquisition of Manus in circumstances where questions arose as to Manus’ continuing links to Chinese technology, talent and operations, despite its Singapore structure. On that basis, the central issue is not whether offshore structuring is lawful in principle. It often is. The issue is whether the structure is supported by real substance and whether it can withstand scrutiny in a sensitive sector.
<span class="news-text_medium">That distinction is important.</span>
Offshore holding structures, Singapore entities, group reorganisations and international capital planning are familiar tools in cross-border technology transactions. Properly used, they may support legitimate objectives, including fundraising, international expansion, investor access and governance efficiency. However, a lawful structure can still create regulatory risk if it is perceived as a template for avoiding scrutiny. Where AI is concerned, perception can be decisive.
The Manus situation therefore highlights a simple point: compliance is not only about whether a structure works on paper. It is also about whether the structure is commercially coherent, operationally real and strategically defensible.
For AI companies, the test is becoming harder. Talent, data, compute, intellectual property, export controls, national security and industrial policy are increasingly interlinked. As a result, the relevant question is no longer simply: “can this be done?”
Instead, the better questions are:
These questions should be asked early, not after public attention has already formed the narrative. The role of lawyers and advisers is therefore not confined to technical execution. It is not enough to identify what is legally possible. Advisers must also consider whether the proposed approach is appropriate, sustainable and capable of being explained in a way that is consistent with regulatory expectations. That requires judgement.
The Manus case also illustrates the risk of public amplification. A company may not intend to become a test case. However, once commentators and market participants present a transaction as a replicable “playbook”, the company may lose control of the narrative. What begins as a private transaction can quickly become a broader regulatory signal. There are four practical lessons.
First, substance matters more than form. Offshore structures must be supported by real governance, management and operational substance. Second, discretion matters. Not every lawful structure should be promoted as a general solution. Third, narrative is also important. How a transaction is described publicly may influence how it is assessed by regulators. Fourth, strategy matters more than arbitrage. Short-term structural optimisation is no substitute for a credible long-term position.
For AI companies expanding internationally, the strongest position is not to obscure their origins. It is to build globally while remaining clear about the company’s technology base, talent pool, governance structure and strategic direction. The next phase of AI globalisation will not reward clever paper structures alone, but those companies that combine innovation with regulatory maturity, substance and disciplined execution.
Manus is therefore best understood not only as a story about one transaction, but as a reminder of the direction of travel. In cross-border AI, regulatory gaps are narrowing. The companies which succeed will be those which understand not only the rules, but also the policy context behind them.
