Tracing the immutable breath of the contract between AI infrastructure and interest rates. The market has long assumed that artificial intelligence will be the great deflationary force—lowering costs, boosting productivity, and giving central banks room to cut. Morgan Stanley’s chief economist Seth Carpenter just flipped that script. His warning: AI may not lead to lower policy rates. In fact, it could push them higher.
The context is critical. We are in a bear market for crypto, but the macro backdrop remains the dominant driver of liquidity. Bitcoin and DeFi protocols live and die by the cost of capital. When rates rise, risk assets bleed. When rates fall, speculative capital flows back into on-chain yield. For months, the prevailing narrative has been that AI will accelerate productivity gains, suppress inflation, and allow the Fed to ease. Morgan Stanley now argues the opposite: AI’s immediate impact will be a massive demand shock—capital expenditure on data centers, chips, and energy grids—that pushes the natural rate (r*) upward. This is not a short-term blip. It is a structural shift in the economy’s equilibrium interest rate.
Let me decode this at the mechanism level. The natural rate of interest is the rate that balances savings and investment in a fully employed economy. If AI triggers a wave of investment—think of it as a new Kondratiev cycle of infrastructure buildout—the demand for capital surges. This lifts r. Central banks, in their attempt to keep the economy from overheating, must set policy rates above that new r. The era of ultra-low rates, already fading, may never return. I’ve seen this pattern before in my forensic autopsies of leveraged DeFi collapses: when the base rate shifts structurally, every levered position gets repriced. The liquidation engines run hotter. The on-chain debt market becomes a minefield.
From my audits of lending protocols on Ethereum and Solana, I can tell you exactly what a sustained high-rate environment does to crypto. First, stablecoin yields—the risk-free rate of DeFi—will remain elevated. That sounds good for lenders, but it crushes borrowers. Leveraged yield farming becomes unprofitable. The demand for leverage drops, and TVL migrates to the safest pools. We saw this in 2022: as the Fed raised rates, Compound and Aave’s utilization rates collapsed. The same dynamic will repeat, but with an added twist: AI-driven demand for capital in traditional markets competes directly with crypto for investors’ dollars. Why take on smart contract risk for 5% when you can buy a T-bill yielding 5.5%? The opportunity cost of holding speculative tokens rises.
The contrarian angle here is uncomfortable. Most crypto natives believe AI and blockchain are symbiotic—decentralized compute, tokenized AI models, autonomous agents trading on-chain. That narrative may be premature. If Morgan Stanley is correct, the first-order effect is not technological convergence but capital competition. AI’s massive CapEx needs will soak up liquidity, leaving less for high-risk, high-uncertainty assets like crypto. Even if crypto’s underlying technology improves, the macro tide is pulling in the opposite direction. Silence in the code speaks louder than audits when the macro environment is the final validator of risk appetites.
Consider the implications for specific sectors. Layer-2 scaling solutions that depend on low-cost throughput may find their growth capped if capital flows toward AI hardware instead of blockchain infrastructure. DeFi protocols with high leverage, like those offering leveraged staking or perpetual swaps, will face thinning liquidity. Stablecoin issuers—especially those backing with real-world assets—will benefit from higher yields, but at the cost of systemic tightening for the rest of the ecosystem. Ethereum’s staking yield, already sensitive to withdrawal volumes, could decline relative to risk-free rates, reducing the appeal of ETH as a yield-bearing asset. The architecture of freedom, compiled in bytes, rests on the shaky foundation of global liquidity conditions.
Where does that leave the crypto investor? The forward-looking judgment is stark: the market is not pricing in a structurally higher rate path driven by AI demand. Positions that rely on aggressive rate cuts in 2025 are vulnerable. Long-duration assets—high-multiple tokens, early-stage protocols without revenue—will be punished. The trade that emerges from this forensic analysis is defensive: prioritize protocols with real cash flows, low dependency on leverage, and strong governance token models that can absorb shocks. Watch the 10-year Treasury yield like a hawk; if it breaks above 4.5% decisively, the macro regime has shifted. AI is not the deflationary savior the market hoped for. It is the demand engine that keeps the Fed hawkish. Decoding this silent language of central bank policy reveals a single truth: in the bear market, survival means respecting the cost of capital.
