Independent risk scores for 50+ enterprise AI vendors. Each vendor is evaluated across five dimensions: security, transparency, compliance, safety, and market stability.
Data encryption, access controls, SOC 2 / ISO 27001 certification, penetration testing, and incident response history.
Model card publication, training data documentation, benchmark disclosures, and limitation acknowledgments.
Alignment with EU AI Act, NIST AI RMF, GDPR, CCPA, and sector-specific regulations.
Red-teaming programs, content filtering, bias evaluation, and responsible disclosure practices.
Funding status, organizational governance, business continuity planning, and vendor lock-in risk.
Our index currently tracks 50+ AI vendors across foundation model providers, cloud AI platforms, and specialized AI companies. Detailed vendor profiles include historical score trends, certification status, and regulatory alignment.
| Vendor | Category | Overall Score | Risk Tier |
|---|---|---|---|
| Anthropic | Foundation Model | 82/100 | Low Risk |
| OpenAI | Foundation Model | 74/100 | Medium Risk |
| Google DeepMind | Foundation Model | 78/100 | Low Risk |
| Meta AI | Open Source | 68/100 | Medium Risk |
| Mistral AI | Foundation Model | 65/100 | Medium Risk |
| Cohere | Enterprise AI | 71/100 | Low Risk |
Scores are illustrative and updated weekly from the public dataset.
Each vendor is evaluated using 30+ weighted criteria derived from public disclosures, third-party audits, and regulatory filings. Scores range from 0 to 100, with higher scores indicating lower risk.
The complete dataset including historical scores, vendor profiles, and API access is available through our data plans.