Deal Blocked Amid Geopolitical Tensions
The Chinese government has formally ordered Meta to unwind its $2 billion acquisition of AI startup Manus, citing national security concerns. The decision, made on April 27, follows months of scrutiny that began when Chinese regulators announced a review of the deal in January 2026. During the investigation, the two cofounders of Manus, Xiao Hong and Ji Yichao, were instructed not to leave China. This move underscores the escalating US China AI rivalry and the increasing difficulty of cross border tech acquisitions.
Manus Technology and Its Integration into Meta
Manus gained prominence in March 2025 for its general AI agent, which acts as an agentic wrapper around Anthropic’s Claude 3.7 Sonnet. The system uses multiple AI agents, including a planner and an executor, to automate tasks like real estate searches, travel bookings, and even software development. Meta had acquired Manus in December 2025 and had begun integrating its technology into Meta’s Ads Manager, which powers advertising across Facebook, Instagram, Messenger, and WhatsApp. The Manus team had relocated to Meta’s Singapore office, and the company had established a Cayman Islands parent entity to facilitate the acquisition.
Wider Implications for AI Founders
The unwinding of the deal creates significant uncertainty. Manus may lose access to Anthropic’s Claude models, as Anthropic restricts AI sales to entities in China. A former Biden administration national security official noted that if Manus remained a Chinese company, its core product would disappear. The failure of this Singapore washing model, where Chinese founders restructure abroad, signals that future founders must establish operations outside China from day one. This setback also challenges Meta’s AI pivot, which followed an $80 billion investment in the metaverse.
CVE References
No CVEs are directly referenced in this article.
Source: Arstechnica
