Last updated: February 2026Data current as of: February 2026
Data Sources & Quality
The CMC Framework maps 730 AI capabilities across 7 industries, profiling infrastructure requirements across 6 structural levers via a 5-stage mapping pipeline with paired quality gates. Each score carries an explicit confidence rating — high, medium, or requires independent validation — so users know exactly what the data supports and where to verify independently.
Confidence at a Glance
High Confidence:Capability names, gap calculations, operating range rules, transmission formula
Medium Confidence:CMC requirements, industry baselines, vendor fit assessments, AI acceleration rate
Verify Independently:Deployment maturity, vendor product availability, org velocity rates, transmission ratios
High Confidence
- • Capability names and descriptions
- • Gap calculation logic
- • Operating range classification rules
- • Transmission ratio formula
Medium Confidence
- • CMC requirement levels (analytical estimates)
- • Industry baselines (estimated, cross-source variance)
- • AI acceleration rate (composite estimate)
- • Vendor product fit assessments
- • Capability completeness
Requires Independent Validation
- • Deployment maturity classifications (market changes fast)
- • Vendor product availability (products change)
- • Baseline estimates (need real assessment data)
- • Org velocity / level-up rates (context-dependent)
- • Transmission ratios (compound uncertainty from inputs)
How to Cite This Analysis
Stone, J. (2025). "AI Infrastructure Requirements Database: Data Sources & Quality." CMC - Context Modelling Capability. https://contextcapability.com/about/data-quality
Researchers: Contact contact@contextcapability.com for dataset access and methodology documentation.