Anthropic Rolls Out AI Scenario Tools and Economic Policy Framework

On September 9-10, Anthropic co-founder Jack Clark released a series of AI scenario tools and policy proposals to help the public and policymakers prepare for different AI futures. The toolkit includes a scenario model starting from 2026, an interactive "AI futures explorer," and an accompanying Anthropic economic policy framework. Clark cautioned that some of the scenarios are quite aggressive, implying society needs to think seriously about how advancing AI capabilities will reshape the economy.

Confirmed

• Anthropic released the Advanced AI Framework and an Economic Policy Framework, the latter offering US policy recommendations for AI-driven labor market shocks, tiered into three levels by severity of impact
• Anthropic established a $200M fund to support randomized controlled trials (RCTs) on AI employment interventions
• The scenario model runs from 2026; responding to Ethan Mollick, Clark said the exercise focuses on economic modeling and does not yet incorporate policy interventions—the release is meant to spark exactly this discussion, and the team will think further about the policy dimension
• The explorer draws priors from a survey of 10,000 Americans, with adjustable and user-supplied priors generating large combinatorial sets of scenarios; Clark said the team deliberately included upper and lower bounds they considered reasonable, including extreme cases

Why it matters

• The explorer's immediate backdrop is Clark's debate with FT journalist John Burn-Murdoch over whether extreme scenarios are a legitimate way to frame the discussion; Clark defended including extremes by analogy to climate modeling
• Clark proposed "pre-registering" AI economic forecasts: have different communities align their predictions to different scenarios now, then check a year later against actual economic outcomes to see whose forecasting methods perform better—introducing a verifiable accountability mechanism to AI prediction
• The economic policy framework and RCT fund move AI's labor-market impact from discussion into empirically testable policy intervention

2026-09-09 ~ 2026-09-10 · 8 related posts

• [source] Jack Clark releases AI scenario model to help policymakers prepare for radical futures — jackclarkSF · 2026-09-09
• Jack Clark: Our 2026 AI Economic Scenario Is Silent on Policy, We'll Think More — jackclarkSF · 2026-09-09
• [source] Anthropic's economic policy framework backs $200m fund for RCTs on AI labor disruption — jackclarkSF · 2026-09-09
• [source] Anthropic's Jack Clark launches AI futures explorer with priors from 10,000 Americans surveyed — jackclarkSF · 2026-09-10
• Jack Clark explains variable-parameter AI futures explorer and extreme scenario bounds — jackclarkSF · 2026-09-10
• Jack Clark: cross-walk the economy in a year against AI scenario communities' predictions — jackclarkSF · 2026-09-10
• Jack Clark proposes pre-registering AI economy forecasts to score predictors in a year — jackclarkSF · 2026-09-10

1 near-duplicate retellings: jackclarkSF