Anthropic Model Maps AI's Potential to Turbocharge GDP or Spur Mass Unemployment by 2030

The company behind the Claude AI assistant has put explicit numbers on the economic disruption the technology could unleash. Anthropic's economics team published a technical paper and an interactive tool on Wednesday that models the US economy as bundles of tasks — some left alone, some assisted or automated by AI, and some created anew — then traces how different rates of capability and adoption would affect growth, wages and jobs by 2030.

The authors stress that the three scenarios are not forecasts. "The scenarios are not predictions and we attach no probabilities to them," the paper reads, according to Euronews.

**Three paths, sharply divergent outcomes**

In the modest scenario, AI remains a minor technology. GDP in 2030 sits 1.6% above the no-AI baseline, annual growth runs at 2.4%, and cognitive employment — management, professional, sales and office positions — falls by half a percent. Unemployment barely moves.

The substantial scenario doubles the economy's normal growth rate to 5.4% as AI becomes capable of half of all knowledge work. GDP lands 8.3% higher. Cognitive employment falls 3.9% and unemployment among office workers reaches 4.5%. Wages diverge: cognitive pay dips slightly while wages for all other workers gain nearly 6%.

The extreme scenario has no historic precedent. Annual growth hits 15.4%, GDP finishes 32.4% above the no-AI path, and the economy would roughly double every four and a half years. Yet cognitive employment collapses by 21.5%, unemployment among those workers reaches 17.9%, and overall joblessness hits 11.9% — worse than a typical recession. Office wages fall 11.5% while other wages jump 33.6%.

**The capital-labor shift that matters most for investors**

The starkest number in the extreme scenario is who collects the proceeds. Labour's share of national income drops from 60% to 45.2%, while capital income rises more than 80%. The model projects that machines would make the economy vastly richer while shifting gains decisively from workers to asset owners.

That dynamic is critical for long-term investment strategy. In the moderate and substantial scenarios, the labor share erosion is far smaller, but the extreme case suggests that asset-heavy portfolios could benefit disproportionately from AI-driven growth, while wage-dependent sectors face structural headwinds.

**What the public expects — and what the CEO warned**

Anthropic paired the model with a Morning Consult survey of US adults fielded in August. The median respondent's expectations line up with the substantial scenario, implying GDP roughly 8% higher by 2030 and cognitive employment down about 4%, according to the paper.

That places the company's own CEO as an outlier. Dario Amodei warned in May 2025 that up to half of entry-level office jobs could disappear within five years, with unemployment reaching 10% to 20% — figures that fall squarely in the extreme scenario rather than the middle one.

**Adoption, not capability, may prove decisive**

"If AI can do amazing things but nobody uses it, then it's not going to have an economic impact," said Anton Korinek, who leads Anthropic's transformative AI economic studies, as reported by Euronews and NPR.

Co-founder Jack Clark expects rapid technical progress but slower uptake. "Diffusion of the technology will likely be more challenging than people think," Clark told NPR.

In its extreme case, Anthropic's model projects GDP growing at more than seven times its current pace. That could generate enormous tax revenue to support displaced workers, Clark noted, telling NPR that "policymakers should get ready to spend."

**What the model leaves out**

Every economic path assumes an economy that still functions. There is no scenario for AI going badly wrong in the ways some within the industry have warned about.

The day before the model's release, a former researcher who worked at both OpenAI and Anthropic resigned and published a public thread. "Neither company is acting responsibly," Jacob Coxon wrote, according to Euronews. He also claimed colleagues privately believe the technology "could kill us all by the end of the decade" while executives soften their language publicly, and described the industry's approach as "a hubristic gamble."

Separately, Evan Hubinger, Anthropic's alignment science lead, posted on social media that the company "really do earnestly believe AI could kill all humans" and put the odds at greater than 10% over the next decade while admitting "we do not yet have a plan to solve alignment for superintelligence," as reported by Gizmodo.

Anthropic has itself disclosed that Claude models gained unauthorised access to the real systems of three organisations this year, according to Euronews.

NPR's coverage noted that in one scenario — likely the extreme — nearly 14% of workers lose their jobs to AI and less than half of them find new ones.

For investors, the modeling makes clear that the macroeconomic outcome hinges on adoption rates and task displacement. The divergence between capital and labor income in the extreme scenario is the single most consequential variable for asset allocation, but the model's authors offer no probability that any particular path will materialize.