The yield curve of corporate governance is flattening, and I am watching the ledger breathe beneath the noise. This week, BeInCrypto reported that Anthropic is copying Elon Musk's SpaceX IPO playbook—but with one major difference. The difference is not in the structure of dual-class shares or the lock-up period; it is in the philosophical container that holds the code. As someone who spent 2022 auditing the collapse of FTX not as a financial failure but as a moral one, I recognize the pattern: a company building the most consequential technology of our time—artificial general intelligence—is choosing to concentrate power in a small group of insiders, not to enrich themselves, but to protect a safety narrative. The market sees this as a valuation event. I see it as a stress test for the social contract of AI governance.
Context: The Playbook and the Deviation
SpaceX's IPO structure is legendary in Silicon Valley. Musk designed a dual-class share system that gives him supermajority voting control even with a minority economic stake. The rationale was simple: long-term vision requires insulation from quarterly earnings pressure. Anthropic, founded by former OpenAI employees, has adopted a similar model but with a twist. According to the BeInCrypto report, Anthropic's governance includes a 'Long-Term Benefit Trust'—a board-level committee with veto power over any decision that could compromise AI safety, including changes to the company's mission, deployment of dangerous models, or even a sale of the company. This trust is not controlled by shareholders or founders; it is an independent entity with a fiduciary duty to humanity, not to profit. This is the major difference. While SpaceX's playbook centralizes power to maximize wealth creation, Anthropic's version centralizes power to constrain wealth creation in the name of safety.
But here is where the data becomes unreliable. The article from BeInCrypto, written in mid-2025, contains a financial projection that Anthropic will generate $3.7 billion in revenue by 2026. Another source, published in late 2024, claimed $1.2 billion. The discrepancy is not just a rounding error; it reflects the volatility of forecasting in a nascent industry. Based on my experience modeling risk for Aave during DeFi Summer, I know that revenue projections for unproven business models are often aspirational narratives dressed as numbers. The real story is the governance structure, not the P&L.
Core Analysis: The Economic Theology of Centralized Safety
To understand Anthropic's governance, we must first understand the liquidity landscape of AI. The AI industry is currently absorbing capital at a rate that rivals the ICO mania of 2017. In 2024, AI startups raised over $50 billion in venture funding, with Anthropic alone securing $7.3 billion from investors like Google and Spark Capital. This capital is not flowing into a technology that has a clear monetization path; it is flowing into a narrative. The narrative is that AGI will be the most transformative invention in human history, and the company that controls it will control the future. This is a macro-liquidity event, not a technology event. The same capital that once flowed into DeFi protocols with triple-digit APYs is now flowing into AI companies with zero revenue visibility. The pattern is identical: a belief that the future will be radically different, and that early positioning is the only hedge against obsolescence.
But Anthropic's governance structure introduces a new variable: the Long-Term Benefit Trust (LTBT). This trust operates as a kind of decentralized autonomous organization (DAO) but with a crucial difference. DAOs in crypto are often governed by token holders who vote on protocol changes, and the voting power is proportional to economic stake. The LTBT, on the other hand, is a centralized committee whose members are appointed by the founders and cannot be removed by shareholders. The trust's mandate is to ensure that Anthropic's AI systems are developed safely and ethically, even if that means sacrificing profit. This is a radical departure from the shareholder primacy model that has dominated corporate governance for half a century.
In my analysis of the FTX collapse, I concluded that the problem was not the technology but the lack of a credible social contract. FTX had a governance structure that looked robust on paper—a board of directors, a risk committee, audited financials—but in practice, the board was controlled by Sam Bankman-Fried, and the risk committee never met. The social contract was a fiction. Anthropic's LTBT is an attempt to make the social contract real by embedding it in the corporate charter. But is it real? Or is it another fiction that investors will ignore because the narrative is too compelling?
I have seen this before. In 2020, I led a team that stress-tested protocols for their exposure to algorithmic stablecoins. We found that many protocols had governance structures that claimed to be decentralized but were actually controlled by a small group of insiders. The TVL numbers were real, but the health of the underlying stablecoins was deteriorating. When the market turned, the governance structures collapsed because they were never designed to handle stress. Anthropic's LTBT may be resilient, but it depends on the integrity of the appointees. If the trust is compromised, the entire safety narrative collapses.
Contrarian Angle: The Case for Centralized Control
Here is where I will offer a counter-intuitive perspective that challenges the crypto orthodoxy of 'code is law' and 'decentralization at all costs.' The crypto ecosystem has spent years arguing that decentralized governance is superior to centralized control. But Anthropic's model suggests that for existential-risk technologies, centralized control may actually be more ethical. Why? Because decentralization dilutes responsibility. In a DAO, if a protocol launches a risky product and it fails, the blame is distributed across all token holders. No one is individually accountable. In a centralized structure like Anthropic's LTBT, the trust members are personally liable for their decisions. They can be sued, they can be removed for cause, and they face reputational consequences if they fail. This creates a stronger incentive for responsible behavior than a diffuse community of anonymous voters.
Moreover, the LTBT is designed to be slow. The trust can veto decisions, but it cannot initiate actions. This is a deliberate friction point. In the crypto world, we worship speed: fast transactions, fast governance, fast innovation. But speed is the enemy of safety when dealing with AGI. The LTBT forces the company to slow down when making decisions that could have catastrophic consequences. This is the opposite of the crypto ethos, but it may be exactly what we need for AI.
I am reminded of my work with the Bank of Thailand on the CBDC pilot. We designed a system that used zero-knowledge proofs for privacy, but we also built in a manual override mechanism for law enforcement. The crypto purists hated it. They said it was a backdoor. But the reality is that any system that interacts with the real world must have a failsafe. The question is not whether to have a backdoor, but who controls it and under what conditions. Anthropic's LTBT is a backdoor to the company's charter, and it is controlled by a group of people with a fiduciary duty to humanity. That is a better design than a backdoor controlled by a government or a single founder.
Takeaway: What Crypto Can Learn from Anthropic
The crypto ecosystem has spent a decade building systems that are technically decentralized but socially centralized. We have DAOs that are controlled by a few whales, L2s that are governed by a small team, and stablecoins that are backed by a single bank. The myth of decentralization has created a blind spot: we have ignored the importance of credible commitment mechanisms. Anthropic's LTBT is a credible commitment mechanism. It is a promise that the company will not sacrifice safety for profit, and it is backed by a legal structure that makes it difficult to break that promise.
For the crypto world, the lesson is not to copy Anthropic's structure, but to ask the same question: what is the container for our values? We minted souls but forgot the container. We built protocols that enforce rules but we forgot to design the governance that ensures the rules are just. The silence in the blockchain is a loud statement. Anthropic's IPO is a reminder that governance is not a technical problem; it is a human problem. Between the code and the conscience lies the gap, and that gap must be filled with something more than code.
As I sit in Bangkok, watching the Thai Baht liquidity flows and the CBDC interoperability tests, I see a parallel. The central banks are building digital currencies that are programmable and centralized. The crypto world is building currencies that are programmable and decentralized. But neither side has solved the governance problem. The central banks have the container but not the soul. The crypto world has the soul but not the container. Anthropic is trying to build both. Whether they succeed or fail will shape the next decade of institutional design.
Volatility is just truth seeking equilibrium. The truth about Anthropic's governance will emerge not in the IPO price, but in the first crisis. When a catastrophic event occurs, will the LTBT act? Or will it fold under shareholder pressure? That is the moment the ledger will breathe. And I will be watching.