Meta Bought Manus. Beijing Undid It. Then Manus Raised $500M.

In December 2025, Meta agreed to buy Manus, the AI agent startup, in a deal reported at more than $2 billion. For most founders, that is the finish line. The team signs, the investors get paid, and everyone moves on.

That is not what happened.

In April 2026, China's National Development and Reform Commission ordered the deal unwound. By the summer, Meta had cut Manus off from its internal systems. In August, Manus told users it was an independent company again, and some user data had to be deleted as part of the separation.

Then, on October 8, Manus's parent company, Butterfly Effect, raised more than $500 million in its first round since the breakup. Boyu Capital and IDG Capital led it. Tencent, ZhenFund, and other existing backers joined.

You are probably not selling your company to Meta next quarter. But this story is packed with lessons that apply to a much smaller deal, and even to your first term sheet. A signed deal is not a closed deal, and a closed deal is not always a permanent one.

The timeline matters because each step shows a different kind of risk.

• 2025: Manus launched its agent product through Butterfly Effect, a company founded in China. In mid-2025 it moved its headquarters to Singapore.
• December 2025: Meta agreed to acquire Manus. Reports put the price at more than $2 billion, with some coverage citing about $2.5 billion.
• March 2026: According to reporting cited by Startup Fortune, two senior Manus executives were summoned to Beijing and barred from leaving China while regulators reviewed the sale.
• April 2026: The NDRC ordered the deal unwound, citing foreign investment rules.
• May to June 2026: Meta and Manus separated their systems and stopped sharing data.
• August 11, 2026: Manus announced it would operate independently again. Early investors bought back shares at roughly a $2 billion valuation.
• September 2026: Bloomberg and the Wall Street Journal reported Manus was raising about $500 million at a target valuation near $4 billion.
• October 8, 2026: The round closed at more than $500 million. Tech Startups notes the final valuation was not disclosed.

Moving your company does not erase where it was built.

Critics called the Manus structure "Singapore washing." The idea was that a Singapore headquarters would make the company look less Chinese to a U.S. buyer. Beijing's order made clear that offshore incorporation did not shield a deal when the technology and the team started in China.

A smaller version shows up all the time:

• You built your first prototype while employed somewhere else, and your old employment contract claims ownership of related inventions.
• A university lab helped with early research, and the school has rights to the IP.
• A contractor overseas wrote core code without signing an IP assignment.
• Your company has a foreign subsidiary that holds key code or data.

In each case, a buyer's lawyers will trace where the valuable stuff came from. If the trail is messy, the deal slows down, the price drops, or the deal dies.

What to do now: Make a one-page "where did our IP come from" record. List every person who wrote meaningful code or created core assets, and confirm each one signed an IP assignment to the company. Fix gaps now.

Between signing and closing, a deal can still fail.

Most acquisitions have a gap between the day the agreement is signed and the day money changes hands. During that gap, things must happen: regulatory approvals, shareholder votes, third-party consents, and financing.

The Manus case went further. The deal was announced, integration began, and a regulator still reversed it months later. That is rare. But deals that fail before closing are not rare at all.

When you get an offer, ask these questions before you celebrate:

1. Which approvals does this deal need? Antitrust review, foreign investment review, industry regulators, or customer contract consents.
2. What happens if an approval is denied? Is there a breakup fee, and who pays it?
3. How long is the window between signing and closing? Longer windows mean more chances for something to change.
4. What am I allowed to do during that window? Many agreements restrict hiring, spending, and fundraising until closing.

That last one is the trap. If the deal dies after six months of frozen hiring and no fundraising, you are weaker than when you started. Negotiate room to keep operating.

The reason Manus could raise $500 million after its deal collapsed is that it still had a business.

Before the Meta deal, Manus reported more than $100 million in annualized revenue about eight months after launch, according to The Business Times as cited by Startup Fortune. Tech Startups cites reporting from The Information that it reached about $500 million in annualized run rate by June 2026. Treat both as company or press-reported figures, not audited results.

Either way, investors had something to buy. A team with paying customers and a working product can survive a failed exit. A team that has already moved its customers onto the acquirer's systems may not.

This applies to any partnership that starts to swallow your company, not only an acquisition:

• A big customer that wants you to build only for them.
• A platform partner that wants exclusive distribution.
• A pilot that turns into a months-long integration project with no signed contract.

Each one can be great. Each one can also leave you unable to stand on your own if it ends. If you are mapping out what your company looks like with and without a big deal, a planning tool like Foundra can help you lay out both versions side by side, so you know your runway and revenue if the deal never closes.

Every deal has hidden veto holders. Find them early.

In the Manus case, the veto came from a government that was not a party to the contract. For a typical startup, the veto holders are usually closer to home:

• Investors with protective provisions. Your preferred shareholders may need to approve a sale.
• Key customers with change-of-control clauses. Some contracts let a customer cancel if you are acquired.
• Co-founders or early employees with large stakes. If they refuse, the deal can stall.
• Lenders. Loan agreements often require consent for a sale.
• Regulators. If you touch health data, finance, defense, critical infrastructure, or cross-border data, assume someone will review the deal.

Pull out your cap table, investor agreements, and top customer contracts. Highlight every clause that gives someone else a say over a sale.

Your investors and your location can open some doors and close others.

After the breakup, Manus's new round came largely from China-linked investors, including Boyu Capital, IDG Capital, and Tencent. Coverage in September also noted that Manus was reportedly considering a structure that could support a Hong Kong IPO. As Gadget Review summed it up, domestic and regional capital became the realistic path to liquidity for China-linked AI startups.

For a first-time founder, the lesson is simple: when you take money, you are also choosing who your future buyers might be. Some questions to ask yourself:

• Would a likely acquirer in your market be uncomfortable with a particular investor on your cap table?
• Does any investor bring government ties that could trigger a review in another country?
• If you sell into regulated industries, will your ownership structure pass their vendor checks?

None of this means you should turn down good money. It means you should know what a check might cost you later.

Manus's comeback round reportedly targeted about double the buyback valuation.

It is easy to see a collapsed deal as the end of the road. The Manus story says otherwise. The company returned to the market and raised a very large round, with reports in September pointing to a target valuation near $4 billion, roughly double the buyback price.

There are real costs, though. Manus had to delete some user data. It spent months on separation work instead of product. Its executives faced personal restrictions. And its future options narrowed.

If a deal you were counting on falls apart, here is a short recovery checklist:

1. Tell your team the truth quickly. Rumors are worse than facts.
2. Call your top customers. Confirm you are still here and still shipping.
3. Rebuild your runway plan. Assume the deal money is gone for good.
4. Restart investor conversations. Lead with the business, not the drama.

Why did China block Meta's acquisition of Manus? China's NDRC ordered the deal unwound in April 2026, citing foreign investment rules. Reports said regulators were concerned about AI technology, data, and talent that originated in China moving to a U.S. company, even though Manus had relocated to Singapore.

How much did Manus raise after the deal collapsed? More than $500 million, announced October 8, 2026. Boyu Capital and IDG Capital led the round, with existing investors such as Tencent and ZhenFund participating. The final valuation was not disclosed.

Can a regulator block a small startup acquisition? Yes. Size matters less than subject matter. Deals involving sensitive data, defense, health, finance, or certain technologies can face review even at small dollar amounts. Ask a lawyer early if your company works in any of these areas.

What should I check before signing an acquisition agreement? Required approvals, breakup fees, the gap between signing and closing, your operating limits during that gap, investor approval rights, and customer change-of-control clauses.

Does moving my company's headquarters reduce regulatory risk? Not always. Manus showed that regulators may look at where the team, code, and data originated, not only where the company is incorporated.

1. Manus raised $500 million in its first funding round since China blocked Meta's acquisition, Quartz, October 8, 2026
2. Startup Funding News Today, October 8, 2026: Manus, Universal Quantum, Mecka, Verso & More, Tech Startups
3. Manus is breaking from Meta as China forces the $2 billion AI deal apart, Quartz, August 11, 2026
4. China Forces Meta to Unwind Its 2 Billion Dollar Acquisition of Manus AI, Startup Fortune, August 15, 2026
5. Manus Seeks $4B Valuation After Beijing Killed Meta Deal, Gadget Review via Yahoo Finance, September 18, 2026