NIST AI RMF 1.0 · AI 100-1 + AI 600-1
The four NIST functions, pre-implemented.
NIST AI RMF 1.0 is the gravitational centre of US AI governance - referenced by the CFPB, FDA, SEC, FTC, EEOC, and most sector regulators. Hootleash implements every category and subcategory out of the box, including the Generative AI Profile (AI 600-1).
Core functions
Govern · Map · Measure · Manage.
GOV
Govern
The cross-cutting function. Sets organisational culture, accountability, policy, and oversight that make the other three functions repeatable.
- GOV-1 Policy & culture
- GOV-2 Roles & responsibilities
- GOV-3 Supplier oversight
- GOV-4 Capabilities & training
- GOV-5 Compliance
- GOV-6 Lifecycle stewardship
MAP
Map
Establishes context. Identifies AI capabilities, lifecycle stage, intended use, third-party dependencies, and impacts.
- MAP-1 Context
- MAP-2 AI capability characterisation
- MAP-3 Lifecycle stage
- MAP-4 Impact characterisation
- MAP-5 Third-party considerations
MEAS
Measure
Analyses and tracks AI risk with quantitative and qualitative methods over time.
- MEAS-1 Identify methods
- MEAS-2 Evaluate
- MEAS-3 Monitor performance
- MEAS-4 Feedback & assessment
MAN
Manage
Allocates resources to treat risks, prioritises responses, documents residual risk, and communicates with stakeholders.
- MAN-1 Risk treatment
- MAN-2 Prioritisation
- MAN-3 Incident response
- MAN-4 Communication
Generative AI Profile · AI 600-1
GenAI risks NIST flagged. Controls Hootleash ships.
- CBRN information misuse - content filters and tool boundaries enforced inline.
- Confabulation - citation provenance and grounded-generation evidence captured per response.
- Dangerous, violent, or hateful content - multi-tier safety filters plus reputational scoring.
- Data privacy - inline PII / PHI minimisation and purpose-of-use checks.
- Environmental impact - per-call compute accounting and aggregate carbon evidence.
- Information integrity - watermarking, provenance manifests, and tamper-evident logs.
- IP infringement - license-aware retrieval boundaries and audit-ready training data summaries.
Ready when you are
See your stack against the RMF crosswalk.
We're building Hootleash in private design partnership with a small number of regulated enterprises. If you run autonomous AI in production, get in touch.
Pre-launch · design partner program open · early access 2026