Tracing the gas trail back to the genesis block—except this time, the transaction isn't on-chain. It's inside the governance structure of the AI company that's supposed to be the most safety-conscious in the world.
When Crypto Briefing published its quiet report on Cami Clark's informal advisory role to Anthropic CEO Dario Amodei, the crypto-native outlet likely expected a shrug. Instead, the report exposed something far more interesting than a personnel note: a governance anomaly that deserves forensic attention.
The article's core claim is deceptively simple. Cami Clark, described as a "key advisor" to Amodei, exercises "informal influence" over "strategic decisions and securing key investments." That's it. Two information points, one of which is editorial opinion rather than verifiable fact. Yet within this paucity of data lies a structural question that the AI industry—and the crypto ecosystem watching it—has yet to confront.
The Context: A Company Built on Governance Theater
Anthropic's entire brand is predicated on governance. The company's Public Benefit Corporation (PBC) structure, its Long-Term Benefit Trust, and its much-publicized "Constitutional AI" framework all serve a single narrative: we are not OpenAI, we will not race recklessly toward AGI.
This is governance as product differentiation. The company's valuation—which has climbed past $60 billion in recent rounds—rests substantially on the trust premium it has extracted from enterprise clients and policymakers. Amazon invested up to $4 billion. Google committed another $2 billion. These are not ordinary venture bets; they are strategic partnerships built on the assumption that Anthropic's governance architecture ensures long-term stability.
The entropy in this system is that formal governance structures and informal influence networks often run on parallel tracks—and the latter frequently determines the former's actual trajectory.
The FTX collapse in late 2022 nearly killed Anthropic. The company had banked on Sam Bankman-Fried's billions, and when those evaporated, it faced an existential funding crisis. That rescue came through Dario Amodei's personal networks. This history matters: Anthropic's survival has always depended on the CEO's ability to activate high-trust relationships at critical moments.
Cami Clark's role sits precisely within this pattern. She is not listed in Anthropic's official governance documentation. She holds no board seat, no executive title, no SEC-disclosed advisory contract. Yet the report suggests she shapes strategic decisions and facilitates capital access.
The Core: When Informal Influence Becomes Systemic
Here's what a standard analysis misses: this is not a gossip item, it's a structural vulnerability.
In the absence of trust, verify everything twice—but when the trust itself is informal, there's no verification mechanism at all.
Consider the investment dimension. Anthropic has raised over $7 billion cumulatively. Its most recent rounds involved complex negotiations with hyperscalers who demanded compute commitments, equity stakes, and governance concessions. In this environment, a trusted intermediary who can navigate between the CEO's vision and investor concerns carries enormous transactional value.
But the same network that enables capital formation creates an accountability gap. If Clark helps secure investments—as the report suggests—she is functionally participating in fiduciary decisions without fiduciary responsibility. She can influence which investors get access, at what valuation, and with what terms, while bearing none of the legal liability that formal officers carry.
The security dimension is even more concerning. Anthropic's "safety-first" positioning isn't just marketing; it has concrete operational implications. The company makes decisions about model release timing, capability restrictions, and alignment research priorities. If informal advisors shape these decisions, the company's formal safety mechanisms become partially decorative.
Smart contracts don't have this problem—their governance is executable code. AI companies still run on human trust, which is why their governance structures need twice the rigor, not half.
The Contrarian Angle: The Blind Spot Nobody Wants to Discuss
Here's the counterintuitive position: informal advisory networks may be a feature, not a bug—at least in the short term.
AI companies face a fundamental tension. They must project institutional stability to attract capital and enterprise customers, while maintaining the agility to pivot quickly as the technology evolves. Formal governance processes are slow. Committees deliberate. Boards approve. Trust structures constrain.
The unofficial advisor model resolves this tension by creating a parallel decision-making channel. Dario Amodei can consult Clark—or any other informal advisor—without triggering board review, investor notification, or public disclosure requirements. This preserves optionality.
But this is precisely the vulnerability. Optimism is a feature, not a bug, until it fails—and when it fails, it fails exactly where the formal governance structure should have caught it.
The 2025 AI landscape is littered with governance failures: board coups, sudden leadership changes, mission drift under investor pressure. OpenAI's Sam Altman saga demonstrated that even the most sophisticated governance structures can be circumvented when personal networks override institutional processes.
The deeper blind spot is crypto-specific. Crypto Briefing's coverage of Anthropic personnel matters suggests a convergence that few market participants have fully processed: the AI capital formation ecosystem is increasingly overlapping with crypto-native capital. If Clark has connections to crypto/Web3 investment circles—which her coverage on a crypto outlet hints at—Anthropic may be accessing capital sources that carry different risk profiles, regulatory exposure, and expectations than traditional tech investors.
The Takeaway: Watch the Governance, Not the Headlines
Entropy increases, but the invariant holds—and for Anthropic, the invariant is that personal networks, not formal structures, determine strategic outcomes.
For those tracking AI-crypto convergence, the Clark report is a signal. It suggests that the next major AI capital raises may increasingly flow through informal channels, with implications for transparency, accountability, and investor protection.
The questions worth tracking are concrete. Will Clark appear in Anthropic's next SEC filing? Will the company formalize its advisory structures as it matures? Or will it maintain its current shadow governance model?