Global
CCC
High exposure, unevenly demonstrated governance
525 assessed · 498 rated
REI 35.2 · ADI 20.6 · T 69.0%
150 ratings at T≥90Weighted monitored field: CC
The headline bands show comparative position. REI shows exposure, ADI demonstrated governance, and T the share of the evidence surface that can be reconstructed from public sources.
| Industry | n | Band | REI | ADI |
|---|---|---|---|---|
| Healthcare | 103 | B | 39.1 | 31.4 |
| Public administration & benefits | 49 | B | 44.7 | 26.0 |
| Migration & border | 12 | B | 36.7 | 32.0 |
| Models served as a service | 26 | CCC | 50.2 | 19.2 |
| Platforms & content | 65 | CC | 27.4 | 21.0 |
| Finance & credit | 55 | CC | 28.0 | 17.8 |
| Foundation models | 35 | C | 36.7 | 10.6 |
| Consumer creation & media | 61 | C | 30.5 | 8.3 |
A RATE AI band is a comparative market position. Its value is knowing what the position means for commitment, evidence, continuity and the next decision - while there is still room to act.
Evidence and control capacity support a stronger comparative position. Keep the evidence current and preserve the conditions behind consequential decisions.
The position remains actionable while evidence, ownership or exposure gaps merit active closure before the next material system, market or procurement change.
The position calls for focused executive attention to evidence, ownership, redesign and decision conditions while corrective choices remain available.
The observed position reflects substantial pressure on defensibility, market access or operating continuity. Evidence preservation, containment and structured response become immediate priorities.
Global, jurisdiction, industry and sector reports translate the completed GEN1 field into commercial and institutional decision intelligence.
The global report connects external proof, regulatory timing, enterprise service-layer exposure, provider dependence and decision defensibility across 525 assessed AI systems.
Read the global report →The statistics are anchors. The analysis is about what they change for market access, procurement, regulatory readiness, executive accountability and the ability to keep moving.

150 of 498 ratings sit at T≥90, where the position is point-wise checkable against the public record. Across 278 systems with partially visible governance evidence, a typical system is around two concrete governance records away from a stronger, more independently verifiable position.
Healthcare leads the complete large-industry field at B. Of 103 assessed systems, 51 positions sit at T≥90 and 12 are in the A zone. The advantage aligns with authorisation infrastructure that leaves durable, reusable evidence behind.

122 of 525 AI systems assessed are high-risk under the frozen AI Act benchmark while operating under no binding risk class in force. Twenty-six also show zero publicly demonstrated controls. The gap is a measurable prioritisation map for regulators, boards and investors.

Public AI resolves to B across 111 systems while private AI resolves to CC across 414. The pattern changes by market: private systems lead the public subset in North America, while the UK shows a decisive public advantage and the EU a smaller public lead. Ownership matters less than the evidence infrastructure around the decision.
The public rating shows where the field sits. Independent Human Oversight, Decision Traceability and Dependency Resilience bring that external position into live AI operations.
How the industry and jurisdiction environment is moving.
What the evidence supports for the defined launch or scale decision, with conditions and review triggers.
Which assurance patterns are visible across the relevant field.
What provider, model and evidence dependencies mean for the procurement decision and contract conditions.
How legal clocks and public-evidence expectations differ across jurisdictions.
What evidence, controls and responsibilities change for the defined deployment.
What the public band says about the external position.
What was known, who owned the decision, what changed and which evidence and response route apply.
Collection is continuous. Public aggregate editions follow a quarterly cycle, with the next editions scheduled for December 2026; client reporting runs monthly, on real-time monitoring. RATE AI analyses are cross-read against current official policy and regulatory signals, including EU AI Act implementation, OECD lifecycle accountability, sector regulators, public-sector transparency standards and national AI strategies.
Scoring logic, evidence discipline, component analysis and publication rules remain available for technical readers. They provide the technical foundation for the public interpretation.