Guide · Governance
AI Readiness
AI readiness describes whether people, processes, and systems are prepared to use AI responsibly. It is broader than AI production readiness, which asks whether a specific AI application can safely operate in production.
AI readiness is broader than APRF’s production gates—use the framework when you need measurable pass/fail controls for a live system. Start here: AI Production Readiness Framework
What AI readiness means
AI readiness is the umbrella term teams use for preparedness: skills, governance, data quality, risk appetite, tooling, and operating model. It answers questions like “Are we ready to adopt AI?” or “Is our organization mature enough to run AI programs?”
That is useful—and incomplete for shipping. Many organizations score well on strategy workshops while their customer-facing assistants still lack spend ceilings, tool allowlists, evaluation gates, or incident playbooks.
AI readiness vs AI production readiness
| Question | AI readiness | AI production readiness | | --- | --- | --- | | Scope | Org, program, or portfolio | A specific AI application or agent system | | Typical signal | Policies, training, pilots, roadmap | Measurable engineering controls that pass or fail | | Failure mode | Slow adoption, shadow AI | Outages, abuse, data leakage, runaway cost |
AI production readiness narrows the question to: *Can this AI application safely operate in production?* That is the focus of the AI Production Readiness Framework (APRF).
How to assess AI readiness without a vanity score
1. Separate program readiness (ownership, risk acceptance, compliance mapping) from system readiness (security, evaluation, operability, cost).
2. For each production AI system, apply a gated profile—mandatory checks with artifacts and pass conditions—not a single percentage badge.
3. Cite a pinned framework version when you claim readiness so results stay comparable over time.
APRF’s Core Profile is the production-minimum gate set for customer-facing systems. Regulated / Tier‑3 systems need the stronger profile. Self-attestation quizzes are not third-party certification.
Where to go next
- AI production readiness — the production-scoped definition and gate model.
- AI Production Readiness Framework — homepage for APRF (spec, domains, stewardship).
- How APRF works — maturity, criticality, lenses, and crosswalks.
- Core Profile — normative mandatory checks.
Next: AI Production Readiness Framework
Open the related pillar specification for mandatory checks, artifacts, and pass conditions. Self-attest is optional.
Related
Frequently asked questions
- What is AI readiness?
- AI readiness is organizational and system preparedness to adopt and operate AI—covering people, process, governance, data, and tooling. It is broader than whether one AI application is safe to run in production.
- Is AI readiness the same as AI production readiness?
- No. AI readiness is the umbrella. AI production readiness asks whether a specific AI application can safely operate in production using measurable engineering controls.
- How does the AI Production Readiness Framework relate to AI readiness?
- The AI Production Readiness Framework (APRF) is a working-draft specification that measures AI production readiness with domains, checks, and gated profiles. It does not replace NIST AI RMF or ISO/IEC 42001; it focuses on production engineering gates.