Fault Line provides a transparent, evidence-based assessment of structural vulnerabilities facing frontier AI laboratories. Each lab receives a Fragility Score from 0–10, where higher scores indicate greater systemic fragility. Scores are derived from a checklist of binary indicators across six dimensions—compute dependence, cloud concentration, policy exposure, demand risk, talent & governance, and resilience—each supported by publicly verifiable news events. A confidence-weighted variant is shown alongside, where low-confidence events carry 0.5× weight and medium-confidence events carry 0.75×.