The CNN town hall in Des Moines, Iowa, on January 10, 2025, lasted 72 minutes. For 71 of those minutes, Donald Trump talked about trade, immigration, and the fading memory of the 2020 election. Then, in the closing minute, he said: "Artificial intelligence is bigger than the internet. And we will build the power plants and data centers faster than anyone. Light touch, no heavy hand. We are going to win this race."
That one minute, parsed by every trading desk in New York, London, and Singapore, set off a ripple that reached the crypto markets by the next morning. Bitcoin futures on CME spiked 2.3% in two hours. The CoinDesk AI Index, a basket of tokens tied to decentralized compute, rose 4.7%. The market interpreted the signal as a macro tailwind for any asset that consumes energy or rides the wave of American technological exceptionalism.
But the signal is not that simple. When a politician promises "light touch" regulation on AI, the subtext for crypto is a regulatory philosophy that tolerates existential risk in exchange for speed. The same philosophy, applied to digital assets, would mean no SEC enforcement, no Treasury sanction lists, and no CFTC rules on stablecoins. In theory, that is bullish. In practice, the infrastructure required to satisfy AI—new power plants, massive data centers, and high-voltage transmission lines—will compete directly with the infrastructure that Bitcoin and Ethereum need to survive. The grid does not scale infinitely. The capital does not flow to two masters.
Context: The Global Liquidity Map and the AI Energy Draw
To understand the impact, one must first map the global liquidity flows that have shaped crypto since 2020. The post-COVID era saw trillions of dollars printed by central banks, a portion of which flowed into alternative assets through the "risk-on" channel. Bitcoin’s correlation with the Nasdaq 100 peaked at 0.72 in 2021. That correlation has since decayed to 0.35, but the underlying driver—liquidity—remains the same.
Now, a new liquidity sink is emerging: AI infrastructure. Goldman Sachs estimates that US data center power demand will grow by 15% annually through 2030, requiring $1.2 trillion in cumulative investment. The Department of Energy projects that new AI data centers will require 50 gigawatts of additional capacity by 2028, equivalent to the entire current grid capacity of Spain. This is not a marginal shift. It is a structural reallocation of capital and energy.
Crypto’s energy consumption, while significant, is dwarfed by this coming wave. Bitcoin’s annualized electricity consumption sits at around 150 terawatt-hours, roughly the same as the Netherlands. AI data centers, if built at the pace Trump envisions, could consume 400 terawatt-hours by 2028. The two are not apples-to-apples, but they are both competing for the same limited grid capacity, the same regulatory approvals, and the same pool of institutional capital.
Core: Crypto as a Macro Asset in the AI Infrastructure Race
At first glance, a pro-AI, pro-infrastructure policy should be a tailwind for crypto. The reasoning is straightforward: more data centers means more demand for low-cost, reliable energy. Bitcoin miners are the most flexible energy consumers on the grid. They can curtail operations in seconds, providing grid stability, and they can locate in stranded energy zones where utility-scale data centers cannot. In a world where energy is the new gold, Bitcoin miners become the swing producers.
But the nuance is in the political economy. Trump’s "light touch" language extends to environmental reviews. He wants to bypass the National Environmental Policy Act (NEPA) for data center construction. If that happens, the regulatory advantage that Bitcoin miners have—the ability to use flared gas, hydropower, or curtailed renewables without lengthy permits—evaporates. Traditional data centers, with their enormous cooling needs and 24/7 uptime requirements, will suddenly be able to build anywhere. The grid interconnection queue, already backlogged with 2,000 gigawatts of projects, will favor the politically connected hyperscalers over the decentralized miners.
I experienced a similar dynamic during the 2020 DeFi summer. While auditing Aave v2’s liquidity flows, I noticed that stablecoin issuers were funneling capital into yield farms that paid 100% APY, but the underlying collateral was concentrated in a single bridge. The analogy here is energy concentration. If AI infrastructure absorbs all the new capacity, Bitcoin’s hash rate growth will slow, and the security model that underpins the entire crypto ecosystem—proof-of-work—will become more expensive. The "light touch" for AI becomes a heavy hand for crypto.
Contrarian: The Decoupling Thesis—Crypto as a Counter-Cyclical Hedge
The consensus view, reflected in the 4.7% jump in the AI token index, is that Trump’s pro-AI stance is pro-crypto. I disagree. The decoupling thesis I have been tracking since 2023 suggests that crypto will not benefit from the AI infrastructure boom, and may actually suffer from it.
Consider the capital flows. Institutions that allocate to AI are the same institutions that allocate to crypto. BlackRock, Fidelity, and Goldman Sachs are the common denominators. The spot Bitcoin ETF inflows in 2024 totaled $35 billion, a record. But those inflows came from a pool of institutional capital that was also pouring into AI stocks. If AI infrastructure requires $1.2 trillion in investment, the opportunity cost for crypto allocation becomes acute. The marginal dollar will go to the asset with clearer regulatory backing and a more direct narrative of American technological dominance. Crypto, still a "wild west" in the eyes of risk committees, loses.
Moreover, the "light touch" regulatory philosophy may not apply to crypto. Trump’s team has made no mention of digital assets. The only crypto-related signal in his campaign has been a vague promise to "stop the war on crypto" by firing SEC Chair Gary Gensler. But that is a political statement, not a policy framework. The same politicians who support AI deregulation often support stricter crypto regulation, because they see AI as a tool for national security and crypto as a threat to dollar dominance.
I recall the NFT mania in 2021, when I invested $20,000 in CryptoPunks to understand the shift from utility to social signaling. The lesson I learned was that narratives can decouple from fundamentals. The AI narrative has decoupled from crypto. The market is pricing in a correlation that may not exist.
Takeaway: Positioning for the 2025 Cycle
If Trump’s AI policy becomes law, the winners in crypto will not be the broad market. They will be specific niches: tokens that represent compute resources (like Akash, Render, or Golem), projects that provide energy flexibility (like Bitcoin miners with demand response contracts), and protocols that facilitate decentralized AI inference (like Bittensor or Fetch.ai). The broad market, especially proof-of-work chains, will face headwinds from energy competition and regulatory neglect.
My advice is to position for a bifurcation. Go long on decentralized compute and AI-infrastructure tokens that benefit from the energy buildout. Go short or neutral on layer-1 tokens that are energy-intensive and have no AI angle. The macro cycle is turning, but the wind is shifting direction. The question is not whether crypto will survive the AI boom. It is whether crypto will have the energy to run.
