The AI Vendor Risk Score (AVRS) distills complex vendor behavior into a single 0–100 score. Lower scores mean higher business risk. Updated daily with real data.
AVRS combines cost volatility, operational stability, policy drift, model lifecycle risk, and market concentration into a single score.
Consistent pricing, excellent uptime, stable API policies
Recent ToS updates require monitoring, otherwise stable
OAuth policy changes, Claude pricing volatility
Significant rate limit cuts, inconsistent API policies
Model deprecations with short notice, pricing instability
Score Bands: 90–100 Stable · 75–89 Watch · 50–74 Elevated Risk · 0–49 High Risk
Scores updated daily at 06:00 UTC. Based on 12 months of historical data.
AVRS isn't arbitrary. Each score is computed from real historical data across five weighted risk factors.
Pricing stability over 12 months. Measures frequency and magnitude of price changes.
API uptime, latency consistency, and error rate stability over time.
Terms of Service change frequency and impact on existing integrations.
Deprecation notice periods, migration support, and model stability patterns.
Ecosystem dependency risk and competitive positioning factors.
Consistent pricing, reliable service, stable policies. Safe for critical workloads.
Generally stable but recent changes warrant monitoring. Suitable for most use cases.
Higher volatility or policy instability. Consider redundancy for critical systems.
Significant instability or frequent disruptive changes. Minimize dependency.