For decades, the narrative of technology has been one of liberation—code that runs anywhere, markets that need no permission, communities that self-govern. Last week, that narrative collided with a hard reality: Meta offered $2 billion for Manus, a promising AI startup, and Tencent—a company with no official stake in Manus—quietly led a campaign to unwind the deal. The acquisition collapsed before it could be announced. In the quiet spaces between corporate boardrooms, a decision was made that will shape the future of AI development, and not a single token holder or DAO member was consulted.
This is not merely a story of geopolitical maneuvering. It is a governance autopsy. As someone who has spent the last seven years designing DAO structures and auditing the ethical integrity of on-chain systems, I see in this event a mirror of the very problems we in the blockchain space claim to solve: the concentration of power, the opacity of decision-making, and the absence of consent from those who will inherit the consequences.
Let me step back and frame the context. Manus was not a household name, but within the machine learning community it was known for cutting-edge work in multi-modal agents and applied reinforcement learning. Its technology could integrate with social platforms, VR environments, and autonomous systems—exactly the type of capability that aligns with Meta's vision of an AI-powered metaverse. Meta's offer of $2 billion was generous, roughly 40x revenue estimates, signaling a strategic imperative. Tencent, meanwhile, has its own AI ambitions through the Hunyuan model and sees any outflow of Chinese AI talent to a U.S. competitor as a threat to national competitiveness. By leveraging its extensive network of capital ties and government relationships, Tencent exerted pressure on the deal's stakeholders—venture capitalists, legal advisors, and even the founders themselves—until the transaction became untenable.
From a blockchain perspective, this is a textbook case of governance failure. The decision to sell an asset that will influence millions of users was made by a small, non-transparent group: Meta's executive committee, Tencent's strategic investments team, and Manus's board. No community vote, no on-chain referendum, no token-weighted consensus. The very idea of decentralization—that power should be distributed and decisions should be auditable—was absent.
I have seen this pattern before. In 2017, during the ICO mania, I audited the smart contracts of a project called EtherTrust. They had raised $2 million in a few hours, promising a decentralized lending protocol. What I found inside their code was a hidden admin key that allowed the founders to freeze all funds and drain them at will. When I refused to sign off on their audit, they called me a 'blocker' and tried to push the code live anyway. I published a whitepaper titled Code as Conscience, arguing that decentralization requires moral accountability, not just mathematical trust. That early clash taught me that the biggest threat to decentralization is not malicious code—it's the human tendency to centralize power when the stakes are high. The Manus deal is that same tendency, now playing out in the AI industry with billions at risk.
But let me dig deeper into the core analysis. The central question is: what governance mechanism could have prevented this unilateral block while still allowing Manus to achieve its mission? A naive answer would be a DAO that owns Manus's IP and allows token holders to vote on any sale. However, as a grounded realist who has seen DAOs fail in practice, I must ask harder questions. In 2020, I joined the Community DAO as its lead governance architect, designing a quadratic voting system to prevent whale dominance. We believed we had built the perfect democratic machine. Then a signature replay attack drained $50,000 from our treasury. The aftermath was not a heroic recovery but a three-month retreat into the Victorian bushlands, where I confronted the fragility of human trust in digital systems. I realized that governance is not a smart contract; it is a social contract. No amount of code can enforce ethical behavior if the participants lack integrity.
So what would a practical blockchain-governed Manus look like? It would need a multi-layered structure. First, a foundation with a non-profit charter that explicitly prohibits unilateral acquisition without a supermajority vote from all stakeholders—users, developers, and token holders. Second, a lockup mechanism that prevents any single entity from acquiring more than 5% of governance tokens in a twelve-month period, modeled on the concept of 'gradual vesting' adopted by some DAOs. Third, an emergency brake that can be triggered by a decentralized autonomous security council to halt any suspicious transfer of IP—this council would be elected annually by the community and include members from different jurisdictions to avoid capture by any single state.
In 2021, I applied these principles when I partnered with indigenous Australian artists to mint 100 NFTs on Ethereum. We ensured that 10% of royalties went directly to community trusts, and we resisted immense pressure from speculators who wanted to flip the assets for quick profit. The project raised $150,000 and preserved the cultural integrity of the collection. That experience confirmed my belief that blockchain's true value lies in preserving human stories and consent, not just speculating on digital scarcity. The Manus scenario is no different: the technology is a vessel for human agency, and when that vessel is sold without consent, it betrays the very purpose of decentralization.
But the contrarian angle is necessary here. We must test our idealism against reality. The standard blockchain response to the Manus situation would be to advocate for full DAO control. I disagree—for three reasons. First, DAOs are notoriously slow at decision-making. The Manus founders needed to respond quickly to market shifts and regulatory changes; a DAO vote on every strategic pivot would have paralyzed the company. Second, DAOs are vulnerable to plutocracy: large token holders (including venture capital funds that often masquerade as 'community') can dominate votes, leading to the same concentration of power we see in traditional boardrooms. Third, in the current geopolitical climate, a truly decentralized AI DAO would face existential threats from both sides. The U.S. might consider it a violation of export controls, while China might see it as a sovereignty challenge. The practical reality is that no blockchain governance can fully insulate a project from state power—at least not yet.
I know this firsthand. In 2022, after the FTX collapse and the market crash, I suffered severe burnout. I withdrew from all public speaking and spent six months in isolation, writing a private manifesto titled The Myopia of Decentralization. It was later leaked and became controversial. In it, I argued that our community had become dogmatic, blindly celebrating decentralization as an end in itself without acknowledging when centralized efficiency is necessary. That manifesto forced me to confront a hard truth: resilience requires acknowledging darkness, not just celebrating light. The Manus deal was blocked by Tencent not because Tencent is evil, but because it acted in its own self-interest. A DAO would have done the same—it would have blocked the sale if the community's interests aligned against it. The difference is transparency. A DAO's decision would be recorded on-chain, with debates visible to all, and the reasons could be challenged through fork mechanisms. Tencent's decision was made in a black box, with no accountability.
Let’s examine the economic incentives more rigorously. The analysis I performed on this event uses a framework I developed for evaluating governance failures in DeFi protocols. I call it the 'Sovereignty Gap'—the distance between those who bear the consequences of a decision and those who make it. In the Manus case, the sovereignty gap is vast: the decision was made by a handful of corporate executives, yet the consequences will be borne by the entire AI ecosystem—users, developers, future generations. A properly designed DAO would close that gap by tokenizing ownership of the AI model's inference rights and allowing continuous voting on strategic transactions. The key insight is that ownership must be dynamic, not static: as the AI evolves, so should the distribution of power over its future.
In my advisory work for a major Australian pension fund in 2024, I negotiated a clause that 5% of their crypto allocation would be directed toward open-source infrastructure projects. Those projects included decentralized governance tools that could someday be used to manage AI assets. The critics called it unorthodox, but it demonstrated that institutional capital can drive positive change if guided by ethical principles. The Manus event validates that pursuit. It shows that centralized capital is still dominant, but also that it is fragile—one phone call from a competitor can collapse a $2 billion deal. A decentralized governance system, while imperfect, offers a more resilient ownership structure: to steal a DAO, you need to control 51% of tokens, not just persuade CEO.
Now, let’s discuss what the industry must do in response. First, we need to accelerate the development of 'Sovereign AI DAOs' that incorporate jurisdictional shielding—such as legal wrappers in friendly jurisdictions (e.g., Wyoming DAO LLC, Swiss foundations) that can hold IP and enforce on-chain governance. Second, we need to build defense mechanisms against corporate capture, like 'poison pill' smart contracts that automatically dilute any entity attempting to acquire controlling shares without community approval. Third, we need to design dispute resolution systems that handle when a community vote conflicts with local law—perhaps using decentralized arbitration platforms like Kleros or Aragon Court.
I realize that much of this sounds abstract. But let me ground it in a concrete analogy. In 2017, I audited a smart contract for a decentralized exchange that had a backdoor allowing the admin to drain liquidity. The founders argued it was necessary for 'emergency maintenance.' I argued that if you need a backdoor, you are not decentralized. They eventually removed it, and the exchange survived. The Manus scenario is the same pattern: the backdoor is not in code but in governance. Tencent exploited that backdoor because the decision-making structure was centralized. If Manus had been a DAO, the backdoor would not have existed—or if it did, it would have been visible and auditable.
The takeaway is not that blockchain will magically solve all governance problems. It won't. But it does force a level of transparency that makes abuse harder. When the next Manus emerges—and it will, because AI is moving faster than our governance systems—we must be ready with structures that embed consent into code. The alternative is a world where a handful of global gatekeepers decide which AI lives and dies, not based on merit, but on geopolitical loyalty.
I leave you with a question: if Manus had been governed by a DAO, would the outcome have been different? Maybe the sale would have been approved—but it would have been approved by the people who actually use and depend on that technology. That is the difference between feudal obligation and democratic consent. In the end, the blockchain community must rise to this challenge, not by rejecting centralization entirely, but by building bridges between the old world of corporate governance and the new world of digital sovereignty. The Manus paradox is a call to action.