Phase
Data collection and campaign execution in progress.
CAISI
Independent research and operating notes on AI agent governance.
Flagship Research Build
The State of AI Tool Sprawl
Multi-organization measurement of AI tool inventory, approval gaps, privilege surface, and regulatory exposure. This page tracks the live build for the flagship CAISI sprawl study.
On this page
Current status
Deterministic claim lock
Headline values remain unpublished until strict validation gates pass.
Focus
Inventory quality, approval posture, privilege surface, and evidence-readiness across tools in active use.
Publication posture
Research artifacts publish only when threshold and claim checks pass.
What will publish with the report
Research package
Final report PDF, methodology, definitions, study protocol, preregistration, and claims map.
Data package
Aggregated findings, tool inventory, privilege map, regulatory gap matrix, and data dictionary.
Executive assets
Executive summary, methodology one-pager, and stat-card copy for distribution.
Reproducibility artifacts
Run metadata, deterministic query mappings, threshold gate outcomes, and hash manifests.
Available now
From the CAISI blog
Post 2
Your Repository Is the Runtime Contract for Agents
Why AI readiness starts with repo design, deterministic commands, and validation entry points.
Post 10
The AI Engineering Maturity Model
A staged roadmap from prompt-first experiments to governed, evidence-rich autonomy.
Notes on scope
This page tracks a live report build. Claims marked TBD
remain intentionally unpublished until campaign run artifacts and
strict validation confirm deterministic reproducibility.