China has intervened to prevent Meta from acquiring Manus, a Singapore-based AI agent startup, in a deal valued at $2 billion. After months of review, Beijing asked both parties to terminate the transaction.
What happened
Manus, which has founders and roots tied to China but relocated to Singapore, had raised U.S. venture capital and surpassed $100 million in annual recurring revenue (ARR). The company had been widely regarded as a clean example of a China-to-global AI exit: move offshore, take American funding, and sell to a Western platform.
Following a multi-month review, Chinese authorities signaled that the deal should not proceed and urged both sides to end the acquisition.
Why this matters
The decision underlines that Chinese regulators view agent-based AI not as a routine SaaS asset but as strategic infrastructure. Agent AI intersects with coding, research, automation, enterprise workflows, and platform-scale data access and distribution — areas Beijing treats as national-security and economic-priority concerns.
As a result, the era of borderless AI startups may be coming to an end: Chinese-rooted agent companies seeking international exits will likely need not only customers and strong models, but also political clearance to complete transactions.
Practical implications
- Beijing’s intervention may set a precedent for blocking similar deals involving AI firms with Chinese ties.
- Investors and prospective acquirers may reassess risks, particularly for companies working on agent AI.
- Startups may face increased requirements for regulatory and political alignment when pursuing cross-border financing or exits.
The Manus case therefore represents more than a single aborted transaction; it signals broader regulatory and geopolitical shifts affecting the global AI ecosystem.
