In the DeFi winter, we didn't have time for politics. We were too busy watching liquidity pools drain and pretending our impermanent loss was just a "temporary setback." But crypto taught me something that translates directly to what just happened in the AI world: when the rules of the game aren't written yet, the people who write them hold all the cards.
Rishi Sunak updated his LinkedIn. That's the trigger event. Former UK Prime Minister, the man who hosted the world's first AI Safety Summit at Bletchley Park in November 2023, now sits as an advisor to both Microsoft and Anthropic. Two companies. One political network. Zero transparency about what he actually does.
The crypto-native response is obvious: this is a revolving door, regulatory capture in its purest form. But that's the surface read. Let me show you what this actually means for anyone who's ever watched a protocol die because it couldn't navigate the regulatory maze. This isn't about AI safety. It's about who gets to define what "safe" means.
The Context: A Window That's Closing Fast
Let me give you the market structure first, because context matters more than headlines.
The global AI regulatory landscape is in its most formative phase since, well, ever. The EU AI Act passed in August 2024, but the implementing rules are still being drafted. The US is still figuring out what its October 2023 executive order actually means in practice. China has its generative AI measures. And the UK? The UK published a whitepaper in March 2023 that took a deliberately pro-innovation, "we won't over-regulate" stance.
That whitepaper was Sunak's baby. He pushed the "AI safety hub" narrative hard, positioning Britain as the honest broker between the US's laissez-faire approach and the EU's precautionary principle. He signed the Bletchley Declaration. He brought together world leaders, tech CEOs, and academics to talk about existential risk.
And then he lost the election. And now he works for the companies he was supposed to be regulating.
Here's what most people miss: the timing. Sunak isn't joining these companies as some honorary figurehead. He's joining during the narrowest window of regulatory flexibility that the AI industry will ever have. The EU AI Act's general-purpose AI rules are being operationalized right now. The UK is deciding whether to align with the EU or strike its own path. The US Congress is still debating whether to even have a federal AI law.
In crypto terms, this is like joining a protocol right before the governance proposal that determines token emissions. Except instead of emissions, it's the rules for an entire industry.
The policy window is the competitive advantage. Not the model. Not the compute. The ability to shape the rules before they're written.
The Core: Three Layers of a Power Play
Let me break this down the way I'd analyze a smart contract. Layer by layer.
Layer One: The Microsoft-Anthropic Alliance vs. OpenAI
Microsoft has invested roughly $13 billion in OpenAI. It's also invested roughly $13 billion in Anthropic. That's not a typo. The company is playing both sides of the AI war, which is smart but creates a coordination problem. How do you support two competitors without alienating either?
Sunak is the solution to that problem. He's the neutral bridge — a former head of state who can sit in rooms with both teams and translate between their policy needs without being seen as taking sides. In crypto terms, he's the multi-sig signer who doesn't hold any keys but provides the legitimacy that makes the whole arrangement work.
Anthropic's positioning makes this even more interesting. They're the "responsible AI" company — the one that talks about Constitutional AI and public benefit corporations and existential risk. That's their brand. It's also their moat. When you're selling safety to enterprise clients and governments, you need to look like you take safety seriously. A former PM who championed AI safety as a national priority? That's not just an advisor. That's a walking, talking certification of your safety credentials.
Anthropic's valuation is built on trust as much as technology. Sunak is a trust multiplier.
Layer Two: The British Angle
Anthropic opened its European headquarters in London in 2024. That's not an accident. The UK is DeepMind's home. It's got a thriving AI research scene. And it's got a regulatory framework that's still being written.
Sunak's role in shaping that framework — both before and potentially after his appointment — gives Anthropic and Microsoft something no competitor can easily replicate: institutional memory of how UK policy actually gets made. Not just the official channels. The informal networks. The relationships with civil servants who are still drafting the AI rulebook.
In crypto terms, this is like having the guy who wrote the tokenomics audit the protocol before you deploy. You know where the bodies are buried because you buried some of them yourself.
Layer Three: The Global Chessboard
Here's the part that most coverage misses. Sunak isn't just a UK asset. He's a G7 asset. His network extends to Washington, Brussels, Paris, Berlin, Tokyo. When he picks up the phone, people answer — not because he's Prime Minister anymore, but because he was, and because he might be again.
That's the real value proposition. AI regulation is going global. The EU AI Act has extraterritorial reach. The US is pushing its own standards. The UK is trying to position itself as the bridge between both. Having someone who can navigate all three jurisdictions — who knows the key players personally — is worth more than any single technical breakthrough.
Microsoft and Anthropic didn't hire a former PM. They hired a geopolitical router with a direct line to every major regulatory node.
The Contrarian Angle: What the Ethics Outrage Misses
Everyone's going to write the obvious story: revolving door, regulatory capture, democracy undermined. I get it. I've seen the same pattern in crypto a hundred times. Gary Gensler went from teaching blockchain courses to running the SEC. Former CFTC chairs join crypto exchanges. The game is rigged, the narrative goes.
But here's what the outrage misses: the alternative is worse.
Think about it. AI companies are making decisions that affect billions of people. They're deploying technology that could reshape labor markets, information ecosystems, and political processes. If the only people shaping those decisions are engineers and MBAs — people who've never had to answer to voters, never had to justify their choices in a public forum, never been accountable to anyone but their shareholders — is that really better?
The revolving door is ugly. But the alternative — an impenetrable wall between the public and private sectors — means that policymakers make rules without understanding the technology, and technologists build systems without understanding the policy consequences. That's how you get laws like the EU's original AI Act drafts that would have strangled open-source development. That's how you get crypto regulations that treat every token like a security without understanding how decentralized networks actually work.
Sunak's move is a symptom of a deeper structural issue: AI governance is too important to be left to either governments or corporations alone, but we haven't built the institutions to handle their intersection. So we get informal arrangements like this instead. Former leaders become advisors. Advisory boards become shadow policy councils. And the public is left to wonder whose interests are actually being served.
I'm not saying this is good. I'm saying it's inevitable. And pretending it doesn't happen — or that it can be stopped with ethical guidelines — is the real naivety.
The question isn't whether Sunak should advise these companies. It's whether we're building the formal structures that make such informal influence unnecessary.
The Hidden Risks Nobody's Talking About
Let me flag three specific concerns that I haven't seen covered elsewhere.
Risk One: The ACOBA Black Hole
The UK's Advisory Committee on Business Appointments (ACOBA) reviews former ministers' commercial appointments. But here's the thing: ACOBA's recommendations are advisory, not binding. And the committee has been criticized for being toothless in enforcing its own guidance.
If Sunak's appointment didn't go through ACOBA review, or if the review concluded with no restrictions, that's not a sign that everything's fine. It's a sign that the system is designed to let former leaders do exactly this — and call it compliance.
Risk Two: The "Safety" Narrative as Marketing
Anthropic talks about AI safety like it's a religious mission. And maybe it is — for the founders. But corporations don't have beliefs. They have strategies. Hiring a former PM who championed AI safety serves a commercial purpose: it signals to enterprise clients and governments that Anthropic is the "safe" choice, that buying Claude won't get you in regulatory trouble, that this is the company you can trust with sensitive data.
That's not necessarily wrong. But it means the "safety" conversation is being captured by commercial interests. The people who should be asking the hardest questions about AI risk are now on the payroll of the companies building the systems. That doesn't invalidate their concerns. But it does compromise their independence.
Risk Three: The Precedent Effect
Sunak isn't the first. Condoleezza Rice joined OpenAI's board. Malcolm Turnbull joined its advisory council. Nick Clegg runs global affairs at Meta. This is becoming a pattern — and patterns become norms.
The danger is that we're normalizing a specific model: former leaders extract financial value from their policy networks by joining tech companies, and tech companies gain policy influence by hiring former leaders. Both sides benefit. The public loses. Not because any individual is corrupt, but because the cumulative effect is a concentration of power that undermines democratic accountability.
The Takeaway: What This Means for You
I've been through enough market cycles to know that fundamentals matter more than narratives. And the fundamental here is simple: policy influence is becoming the scarcest resource in the AI industry.
For investors, that means the companies with the best government relationships — not just the best models — are the ones to watch. Anthropic's enterprise growth and government contracts are worth monitoring. Microsoft's ability to navigate global AI regulation while juggling two competing AI investments is a test of its strategic execution.
For builders, it means the regulatory landscape is about to get much more complex. The days of shipping AI products without thinking about compliance are ending. The companies that build policy awareness into their products from day one — the way smart DeFi protocols build in compliance checks from launch — will have a structural advantage.
For everyone else? Watch the ACOBA review. Watch whether Sunak actually shows up at policy events. Watch whether the UK's AI legislation starts to look suspiciously favorable to Microsoft and Anthropic's interests. That's where the real story will unfold.
Every crash is just a story that hasn't finished being told. And every revolving door is just a governance gap that hasn't been closed yet. The question isn't whether Sunak should be doing this. It's whether we're going to build the institutions that make this kind of informal influence unnecessary — or keep pretending that ethical guidelines and advisory committees are enough.
I didn't start my career cynical. I learned to be. The market teaches you. And the market is telling me that political capital is the new compute — scarce, valuable, and increasingly controlled by a few players who understand its power.
The question is who else is paying attention.
t saying.
