Tyler Cowen: AI won't move the GDP needle as fast as hoped — half the economy is permanently sluggish

Summary

Tyler Cowen offers a structurally pessimistic view on AI's near-term GDP impact: roughly half the U.S. economy — government, nonprofits, higher education, and large swaths of healthcare — is what he calls "perpetually sluggish," collectively representing around 50% of GDP. AI penetration in those sectors amounts to marginal time savings and leisure substitution, not productivity transformation. Meanwhile, the dynamic sectors that are adopting AI aggressively — software, biomedical startups, advanced manufacturing — become more efficient precisely as they shrink relative to the deadweight of the sluggish half. The net effect is that moving the GDP needle meaningfully is structurally very hard.

Cowen's central forecast is 0.5 percentage points of additional annual GDP growth attributable to AI. He acknowledges that 1.0 percentage point is possible, which would effectively cover U.S. debt-service obligations without requiring inflationary adjustment, but he won't stake that as a base case. Elon Musk's triple-digit growth figures he treats as out of scope.

U.S. Fiscal Position

Cowen's debt outlook is notably grim. His base case involves roughly half a dozen years of approximately 7% inflation over the next ten to fifteen years as the mechanism by which the U.S. deflates its debt burden — not default, not fiscal consolidation, but monetization. He frames this as already partially baked in: Fed independence, he argues, was eroded long before Trump's recent public pressure on Jerome Powell, because sustained high deficits structurally subordinate monetary policy to fiscal necessity. Trump's fiscal expansion, in Cowen's view, does more damage to Fed independence than his rhetoric.

On the positive side, U.S. AI leadership is the primary offset. A 1-percentage-point AI productivity boost would theoretically resolve the debt math entirely, but Cowen's 0.5-point estimate leaves the country on a knife's edge.

Labor Markets and Trade Jobs

Cowen pushes back on the Michael Burry thesis that AI-assisted DIY tools (photograph-a-toilet-for-plumbing-guidance) will hollow out blue-collar trades. He concedes household demand for tradespeople may fall around 20%, but argues net employment in those categories rises due to data center construction, electrification build-out, and emerging-market infrastructure development across Africa and elsewhere.

AI Music and Creative Disruption

Suno is already eliminating professional studio work in the music industry. One country music vocalist — described as near the top of their field in the demo-recording niche — saw income go to zero in 2025 as songwriters switched to AI-generated samples. Cowen notes the technology is not yet compelling for end listeners but is fully sufficient for professional workflow substitution. He expects to listen to AI music recreationally within two years. On IP liability, he draws a historical analogy: the Beatles absorbed Chuck Berry, Buddy Holly, and Carl Perkins without compensation, and the industry continued. He expects AI music to follow the same pattern with no meaningful royalty framework emerging.

Investing, Gold, and Bitcoin

Cowen's personal strategy is diversified buy-and-hold. Given elevated macro risk — specifically citing dollar safe-haven erosion visible in gold and silver's reaction to Trump's Fed comments — he is considering adding gold and silver as tail-risk hedges. He explicitly excludes Bitcoin from that hedge function.

Tariffs Postmortem

The 2025 tariff regime produced no upside on its stated objectives. U.S. manufacturing employment continued to fall, allied relationships degraded, and consumer prices rose — Cowen cites Chilean Rainier cherries moving from $6.99 to $9.99 per pound as a concrete example. His summary: "not as bad as feared, but no gain."

Biomedical and Longevity

Separate from AI, Cowen is bullish on cancer and cardiovascular medicine on a 40-year horizon, independent of AGI assumptions. His view is that people who are relatively young today will largely avoid dying from cancer or heart disease, with those gains materializing progressively over four decades.

Aesthetics and Urban Development

Cowen and Patrick Collison are actively working to raise public awareness around architectural quality, arguing that the prevalence of ugly new construction is primarily a failure of taste and political will rather than economics. He points to Shaker Heights near Cleveland in the 1920s — homes built by individual families, not committees — as evidence that distributed construction can still produce beauty without institutional coordination. He identifies Helsinki and modern Copenhagen as contemporary models. On YIMBY, he argues the movement would gain significantly more traction if it could credibly promise neighborhood aesthetic improvement alongside housing density increases.