Capitol Hill will not produce a federal AI statute this year. That is now the consensus, and the market has already shrugged. The S&P 500 sits near record highs while the world's largest technology jurisdiction formally abdicates rule-making authority over its most consequential industry. I do not trade AI equity. I trade capital flows. And capital flows do not read press releases — they read the price of legal uncertainty.
Here is the anomaly worth your attention. A jurisdiction that cannot legislate its frontier technology exports regulatory authority as surely as it exports semiconductors. The AI bill is dead on arrival for two reasons the reporting treats as trivial: insufficient calendar time and insufficient consensus. Both are symptoms of the same pathology — political polarization has made durable legislation a scarce asset. Scarce assets get repriced. The question for anyone with exposure to digital settlement rails is not whether Brussels fills the void. It is who captures the rent.
Let me establish the mechanism before the market prices it.
The European Union has already run this play. The General Data Protection Regulation, passed in 2016, became the de facto global privacy standard not through enforcement reach but through market access. Any firm serving EU citizens adopted it. The pattern — the Brussels Effect — is now being replicated on two fronts at once. The AI Act, negotiated and nearing final passage, imposes risk-tiered obligations: data governance, documentation, human oversight, and conformity assessment for high-risk systems. In parallel, the Markets in Crypto-Assets regulation, MiCA, has moved from text to transition, making the EU the first major jurisdiction with a comprehensive digital asset rulebook.
Washington, meanwhile, has produced a patchwork. A 2023 executive order — EO 14110 — established reporting requirements for large compute clusters and safety testing. It carries no legislative authorization. Its durability beyond the current administration is an open question precisely because it is administrative, not statutory. Federal agencies fill gaps through enforcement. States write their own proposals. The result is not deregulation. It is fragmentation — and fragmentation is the most expensive regulatory state there is.
Now the core analysis, because this is where the tape is mispriced.
The first-order effect looks bullish and is a trap. No federal AI statute means lower near-term compliance cost for US developers. Faster product cycles, looser data acquisition, fewer mandatory audits. Note who benefits: firms with in-house counsel and compliance infrastructure. Large incumbents absorb fragmented rules the way a large ship absorbs chop. Startups cannot. A multi-state patchwork — California, New York, a dozen pending proposals — forces every AI firm to maintain multiple conformity regimes. That is a fixed cost, and fixed costs are regressive. The vacuum does not level the playing field. It tilts it.
The second-order effect is where the rent moves: certification and conformity assessment. When the EU defines what high-risk means, it also defines who gets to certify compliance. That is the real prize — not the rule text, but the authority to interpret and attest. Standards bodies, notified bodies, and audit firms capture recurring revenue denominated in every firm that wants EU market access. This is a rules rent, and it accrues to the jurisdiction that legislates first. The AI Act does not merely regulate. It exports an assessment layer. Any American firm serving 450 million European consumers pays into it.
I have watched this exact dynamic run on the crypto side. A decade in cross-border settlement taught me that the compliance layer, not the asset layer, determines where value actually settles. When I modeled the stablecoin de-pegging cascade in 2022, the firms that survived were not the ones with the best collateral ratios — they were the ones with the lowest regulatory ambiguity in their settlement jurisdictions. Liquidity does not flee volatility. Liquidity flees legal indeterminacy. MiCA understood this. It converted indeterminacy into a schedule.
The third channel is the one nobody is pricing: the courts as de facto regulator. With no statute, litigation becomes the rule-making instrument. The copyright suits against frontier model developers are the template — individual judicial rulings setting industry norms with zero legislative input. This is precisely the regime crypto spent a decade enduring. The SEC governed digital assets through enforcement actions, not rules, and the result was years of retroactive standard-setting that suppressed institutional capital formation. AI is now inheriting that structure. The price of court-defined regulation is not zero. It is a risk premium expressed as a discount rate — a cost that compounds silently across every multi-year capex commitment.
So the consensus framing — America loses leadership to Europe — is directionally right and analytically lazy. Leadership in what, exactly? The answer determines the trade.
Here is where I part company with the standard narrative. The permissive reading is that the EU wins the rule-making contest and the US cedes the throne. I am skeptical. Rules without settlement rails are advisory. Rail control without rules is still control. The dollar clears the overwhelming majority of global trade and the stablecoin supply. If the EU writes the strictest AI conformity regime but every settlement of the resulting value flows through dollar-denominated rails — tokenized treasuries, US-regulated stablecoins, correspondent banking on the Fed's balance sheet — then Brussels captures the certification rent while Washington retains the monetary rent. Both are rents. They are not equal. The monetary rent is larger by orders of magnitude.
Which means the vacuum may not weaken US positioning so much as redirect it. Where the US cannot legislate, it dollarizes. That is the pattern that should concern Brussels, not reassure it. As fragmented compliance pushes firms toward jurisdictions they can navigate, the path of least legal resistance for digital settlement increasingly runs through US-controlled infrastructure — because the dollar layer is the one place with unambiguous, functioning rules. The AI stalemate is not the headline. The stablecoin and tokenized-treasury expansion underneath it is.
An honest read requires conceding what this vacuum does not cover. Congress is not the only policy instrument. The executive branch already regulates compute through export controls, administered by Commerce, requiring no AI statute. The chip layer is governed regardless of legislative gridlock. So no US AI governance is false. It is governance through trade and administrative fiat, durable only until the next administration. That asymmetry — hard instruments such as export controls and chip restrictions versus soft ones such as executive orders and enforcement — is the actual competitive structure. The AI bill's failure changes the soft layer. It barely touches the hard one.
For anyone positioning across this cycle, the takeaway is structural, not directional. The regulatory vacuum is not a gap in the system. It is now a pricing factor embedded in the discount rate applied to every long-duration technology and digital-asset commitment. The market is treating the absence of US AI law as a bullish deconstraint. Measured against the certification rent, the judicial uncertainty premium, and the compliance fixed costs, I read it as slow margin compression financed by near-term euphoria.
The question that should sit on every desk: when the EU's conformity regime goes live and its provisions bind cross-border operators, will US digital-asset firms internalize it voluntarily — or discover, at the moment of market-access denial, that no federal rule was never the same thing as no law?
