The protocol remembers what the regulators forget. Last week, Anthropic filed plans to issue super-voting shares to its CEO ahead of a potential IPO. Ten votes per share. One person. One vision. The broader market sees this as a governance strategy to insulate long-term strategy from short-term profit pressure. The crypto analyst sees something else: a textbook case of centralization dressed in fiduciary clothing.
Context is everything. Super-voting shares are not new. Google, Facebook, and Snap all used them to concentrate power in founders' hands. But in the AI sector, where the product itself is a black box of weights and biases, the governance structure becomes the only external check on power. Anthropic, positioning itself as the "responsible AI" alternative to OpenAI, now faces a paradox: its governance model is anything but distributed. The IPO prospectus, based on leaked drafts, shows a dual-class structure where the CEO holds 10x voting power on all matters—including board composition, merger approvals, and even the company's interpretation of its own ethics charter.
For the crypto community, this is déjà vu. We have watched the same playbook unfold in DeFi: a founding team launches a token, retains 40% of supply, and calls it "community-owned." The difference is that on-chain, the ownership is auditable. The voting power is transparent. The fork is always an option. Anthropic's super-voting shares are the antithesis of on-chain governance; they are the planned economy of corporate control, where the single party cannot be forked, only replaced by a board vote that the CEO himself controls.
Core Analysis: The Economic Logic of Concentrated Power
Let me be clear: super-voting shares are not inherently evil. In early-stage companies, they protect visionary founders from activist investors who demand quarterly growth over long-term R&D. Anthropic's CEO has argued that without this structure, the company could be forced to prioritize profit over safety. That argument has merit. But the crypto ecosystem has a better model: the DAO with time-locked governance and veto mechanisms. In my experience auditing DAO treasuries for seven different protocols, I've seen that the most resilient organizations are those that distribute power across multiple independent actors with clearly defined exit rights. The key is not just voting power, but the ability to exit—to fork the protocol if the governance fails. Anthropic's shareholders, especially retail investors in the IPO, will have no such exit. They will own economic tokens without governance rights. In crypto, we call that a "wrapped token"—a derivative that grants none of the underlying protocol's control.
Based on my work with the Ethereum Foundation grant in 2019, I learned that governance is not about voting; it's about the credible threat of kicking out the bad actor. In a DAO, if the core team misbehaves, the community can fork. In a super-voting share structure, the only fork is a hostile takeover, which is effectively impossible because the CEO controls the board. The economic incentive, then, is not to align the CEO with shareholders, but to align shareholders with the CEO. That is a one-way street.
Contrarian Angle: The Case for Sovereign Stewardship
Here is the counter-intuitive angle: super-voting shares might be the only way to build a truly long-term AI company. Crisis is just code with a high gas fee. The AI space is moving faster than any regulatory framework—faster than any decentralized governance model can currently handle. A DAO with 10,000 token holders voting on every model update would be a disaster. Speed without direction is just volatility. In that context, a single visionary with a super-vote might be the most efficient governance structure for the next five years. The crypto purist will scream "centralization," but the pragmatist will note that even Bitcoin has a single leading developer in Bitcoin Core. The difference is that Bitcoin's protocol is open source. Anyone can audit it. Anyone can fork it. Anthropic's code is closed. The super-voting shares are a closed governance structure on top of a closed product. That is a double lock.
Takeaway: The Fork is the Ultimate Check
Open source is a promise, not a product. Anthropic's governance strategy is a bet that the market will trust a single person's vision over the collective wisdom of the community. That bet might pay off—the IPO might be a massive success, and the AI sector may follow suit. But for the crypto ecosystem, the lesson is clear: the fork is the only true check on centralized power. As regulators and investors look at Anthropic's model, they should ask: where is the exit? Where is the fork? If the answer is "nowhere," then the governance is not a strategy—it is a prison. Regulation is the friction that forces efficiency, but in this case, the friction is being applied to the wrong side of the equation. The protocol remembers what the regulators forget: that power without accountability is not innovation; it is a high gas fee on human freedom.