AI Governance Starter Pack
An AI use-case registry, risk classifier, and control-mapping system aligned to NIST AI RMF and ISO 42001 — the governance layer your AI deployment is missing.
The pain
Every company is rolling out AI tools and agents. Almost none have an AI inventory, risk classification, or control framework. When the inevitable incident happens — hallucinated decision, biased output, data leakage — there is no governance trail. ISO 42001 and the NIST AI Risk Management Framework are the emerging standards, and the early movers will sell governance services into this gap for the next decade.
What gets built
- AI use-case registry: every AI/ML/LLM deployment cataloged with purpose, data inputs, model details, owner
- Risk classification engine: NIST AI RMF-aligned scoring (validity, reliability, safety, security, accountability, explainability, privacy, fairness)
- Control mapping: links each use case to required controls based on risk tier
- Model card / system card templates: auto-populated from registry data
- Pre-deployment review workflow: risk assessment → control verification → approval → registration
- Incident logging: AI-specific incident taxonomy, root-cause categories, mitigation tracking
- ISO 42001 readiness checklist: gap assessment against the new AI management system standard
Stack
- Same Supabase + Next.js stack as MVP 01
- Configurable YAML rubrics
- Pandoc
- Optional MLflow
Pricing
$15K–$30K initial assessment + $1,500–$3,500/mo managed governance retainer
Effort to ship
3–4 weeks for v1
AI Governance Starter Pack demo is on the roadmap
Want a guided walk-through of what this tool will do, or to be the first pilot client when it ships?
Get in touch