I used to think that the biggest threat to decentralization was bad code. Then I watched the Manus saga unfold. A $2 billion acquisition blocked by a single regulator. A founder's travel restricted for months. A venture capital giant—Benchmark—bought out by a Chinese tech conglomerate. The whole episode felt like a governance nightmare—and it was. But the lesson isn't that regulators are evil, or that Meta is a villain. It's that we built a system where a few entities control the fate of groundbreaking technology, and where the people who actually use the technology—the users, the developers, the community—have no say. This is why we need on-chain governance for AI agents. If you can't trust the code, you can't trust the agent.
Here is what the charts won't tell you: The Manus event is not about AI. It's about trust. It's about who gets to decide what an AI agent can do, who it serves, and how its data flows. It's about the same power dynamics that have plagued crypto since the beginning—centralized control disguised as innovation. Follow the fear, not the chart. The charts show a successful startup finding a path forward. The fear shows a fragile system where a single government decision can upend a $2 billion deal, where a founder can be banned from leaving a country, and where a company's future is decided in private boardrooms, not in public forums.
Context: The Manus Story in Brief
Manus is a Beijing-based AI agent startup that built a product capable of autonomously executing complex tasks—booking flights, writing code, even managing supply chains. In early 2025, Meta approached with a $2 billion acquisition offer. The deal was seen as a way for Meta to leapfrog in the AI agent race, which was heating up with OpenAI's Operator, Anthropic's Computer Use, and Google's Project Mariner. But Chinese regulators intervened. They investigated the transaction, citing national security and data sovereignty concerns. Xiao Hong, the founder, and his co-founder Ji Yichao were ordered to remain in China. The travel restrictions were only lifted months later, after Meta formally withdrew the offer. Then came the restructuring: Benchmark, the Silicon Valley VC that had led Manus's Series B, exited. Tencent, along with ZhenFund and HSG, bought out Benchmark's stake. Tencent became the largest single shareholder, but with less than 50% ownership. Manus will continue to operate independently from Singapore, with Xiao Hong returning there.
This is a story of centralization. A single regulator blocked a deal. A single corporation (Tencent) now holds the largest stake. A single founder endured personal restrictions. And the users? They were never part of the conversation. This is the exact opposite of the decentralized ethos that crypto people like me have been evangelizing for a decade.
Core: The Technical and Values Analysis
1. The Code Integrity Failure
During the 2017 ICO mania, I spent nights auditing the Solidity code of Gnosis Safe. I found 12 critical logic flaws in their multi-signature implementation. I submitted those findings not for bounty, but because I believed that code integrity is the foundation of trust. That experience taught me that decentralization is not just about having many nodes; it's about having verifiable, transparent decision-making processes. The Manus acquisition wasn't a technical failure—it was a governance failure. The decisions about whether to sell, to whom, and at what price were made by a handful of executives and board members. The users—the people who trusted Manus with their tasks, their data, their workflows—had no voice. If Manus had been structured as a DAO, with a token that represented governance rights, the community could have voted on the acquisition. Perhaps they would have rejected it. Perhaps they would have accepted it with conditions. But at least the process would have been transparent and auditable on-chain.
Instead, we got a behind-closed-doors negotiation, a regulatory surprise, and a fire sale of shares. The code of Manus's product—the AI agent itself—might be elegant, but the code of its governance is broken. Code is law, but only if the law is written in smart contracts. In this case, the law was written in regulatory filings and board resolutions, invisible to the public.
2. The Human Cost of Centralized Decisions
In 2020, during DeFi Summer, I watched friends lose their savings when Compound's governance token crashed. I interviewed 30 affected retail users, documenting their trauma. That experience taught me that behind every market event, there are real people with real pain. The Manus situation is no different. Xiao Hong's travel restrictions were not just a bureaucratic inconvenience; they were a personal ordeal. He spent months unable to leave China, uncertain about his company's future, his own freedom, and the fate of the team he built. The employees faced uncertainty: would they be absorbed into Meta? Would they be laid off? Would their equity be worth anything? The users who relied on Manus for their daily workflows faced sudden service uncertainty. All of this could have been avoided if the governance of Manus had been more decentralized, with clear rules for change of control, community voting, and transparent disclosure.
The human cost of centralization is not just financial; it's emotional and psychological. When a single entity controls the switch, everyone else lives in fear of the switch being flipped. In crypto, we call this "trust minimization." The Manus story shows that even in AI, trust minimization is not just a technical choice—it's a moral imperative.
3. Authenticity Through Independence (or the Illusion Thereof)
In 2021, while the NFT market was exploding with profile pictures and speculative mania, I launched a small collective called "On-Chain Diaries." We minted 50 digital artifacts representing our daily interactions with Beijing, grounded in verifiable local events. I manually coded the smart contract to ensure royalties went to local artists. That project was a quiet act of resistance against the commodification of creativity. It proved that blockchain could support authentic, small-scale community expression.
Manus's independence—remaining as a standalone company, not acquired by Meta—is a similar act of resistance. But it's an incomplete one. Tencent is now the largest shareholder, and while the ownership is less than 50%, the influence is real. Tencent's cloud services, enterprise applications, and data infrastructure could easily become the backbone of Manus. Is that independence? Or is it just a different flavor of captivity? The key difference from my NFT project is that On-Chain Diaries had no external investors; the community owned the project. Manus has a corporate investor with deep pockets and deeper integration potential. True independence requires decentralized ownership, not just a change of controller.
4. Resilience Amidst the 2022 Collapse and the Manus Parallel
When Terra-Luna collapsed in 2022, I retreated from social media for three months. I restructured my failing education platform, pivoting from token-education to fundamental economic literacy. I wrote "The Stoic's Guide to Crypto Winter," a raw piece on maintaining intellectual integrity when financial incentives vanish. That period taught me that resilience is built on shared suffering, not just shared gains.
Xiao Hong's resilience is remarkable. He faced a regulatory investigation, travel restrictions, and the collapse of a $2 billion deal. Yet he rebuilt the company's structure, secured new investors, and is now moving to Singapore to continue operating independently. This is the kind of resilience we need in the crypto space—the ability to survive centralized shocks. But resilience alone is not enough. We need to build systems that are antifragile, that become stronger when exposed to stress. A decentralized governance model would have made the Manus situation less stressful to begin with, because the community would have been part of the decision-making process.
5. Ethical AI-Crypto Synthesis: The Verifiable Truth
In 2026, I founded "Verifiable Truth," a platform using zero-knowledge proofs to verify AI training data origins without exposing proprietary information. The goal was to ensure that AI models are not just powerful, but accountable. The Manus event reinforces the need for this kind of synthesis. AI agents like Manus have access to user data, execute tasks autonomously, and make decisions that affect real-world outcomes. Without transparent governance—on-chain, verifiable, and community-driven—these agents become black boxes controlled by a few. The solution is not just technical (ZK-proofs, on-chain data provenance) but also governance-oriented: token-based voting, decentralized autonomous organizations for AI agent management, and smart contracts that enforce user rights.
The contrarian angle: Maybe the regulatory block was actually a good thing for decentralization? It prevented a single corporation (Meta) from absorbing a powerful AI agent that could have become a centralized data silo. However, the solution (Tencent) is still centralized. True decentralization would require the agent to be owned by its users via a token or DAO. But is that feasible? The pragmatism test: DAOs are slow and inefficient. They struggle with coordination, voter apathy, and security risks. The Manus case shows that sometimes centralized intervention is necessary to prevent worse centralization. But the takeaway is not that we should accept the status quo; it's that we need to design systems that make such interventions unnecessary. If Manus had been a decentralized protocol from the start, with a community treasury, a transparent governance process, and a token that gives users ownership, the regulatory block would have been irrelevant. The community would have decided the fate of the project.
The blind spot: Most people think the Manus story is about AI regulation. It's not. It's about governance. It's about who holds the keys. In crypto, we obsess over private keys. But governance keys are even more important. The Manus saga is a reminder that until we decentralize the governance of AI agents, we are just trading one central authority for another.
Takeaway: Forward-Looking Thought
If you can't trust the code, you can't trust the agent. The Manus story is a wake-up call: we need to build AI agents that are not just technically decentralized but also governance-decentralized. The tools exist—smart contracts, DAOs, ZK-proofs, token-based incentives. The will is lacking. But events like this will accelerate the shift. The next time a $2 billion acquisition is proposed for an AI agent, the community should be able to vote on it. The code should enforce the outcome. And the founder should not have to fear a regulator's travel ban. Follow the fear, not the chart. The fear is that without on-chain accountability, every AI agent is a potential centralization trap. The hope is that we can learn from Manus and build a better future.
