Anthropic AI Development Pace Simulator

Policy Scenarios

Frontier Parameters

Compute Growth (x FLOPs/yr) 2.5x

Safety Research Allocation 35%

Governance & Verification Cap 0.80

Baseline Capability (Index 0-100) 78.5

10-Year Frontier Trajectory Model (D3 Engine) Year 0 to Year 10 Projections

Raw Frontier Capability

Alignment & Control Maturity

Cumulative Catastrophic Risk Exposure

Strategic Impact Analysis

Projected Risk Score (10-Yr)

14.2%

Catastrophic failure probability under governance

Economic Delay Index

2.1 yrs

Delay in frontier GDP transformation

Safety Maturity Timeline

4.8 yrs

Time to robust alignment verification

Recommended Action

Immediate Slowdown & Verification

Slowdown coordinates verification before capability outpaces safety.

Anthropic AI Development Pace Simulator

Your browser does not support video.

Read the explanation

The simulator evaluates 10-year AI trajectories by comparing frontier capability growth against alignment safety maturity under strict governance. Cumulative catastrophic risk accumulates whenever capability exceeds alignment safety, penalized exponentially by the capability gap. Selecting Business As Usual unleashes compute growth to four point eight times per year, widening the hazard gap and driving catastrophic risk above eighty percent. Restoring Anthropic's Slowdown preset caps compute scaling and raises safety allocation, synchronizing verification and compressing ten-year risk to fourteen percent.