Scope & Mobilize
Kickoff, scoping interviews, and a target-state workshop so you choose where you want to be. Data-access agreements, a communications plan, and the assessment calendar.
The audit is a structured, evidence-based look at how AI is really being used in your organization, what risk that creates, and how ready your leadership, workforce, policies, and controls are to manage it. It ends with a maturity baseline and a risk-weighted roadmap, and that roadmap is what decides which implementation steps you need.
Every audit follows the same four phases, scaled to your size and complexity. A focused single-area assessment compresses to two to three weeks. Any critical legal or regulatory exposure found along the way is escalated to your sponsor immediately, not held for the final report.
Kickoff, scoping interviews, and a target-state workshop so you choose where you want to be. Data-access agreements, a communications plan, and the assessment calendar.
Document review, leadership interviews, workforce surveys, Shadow AI discovery, control walkthroughs, and system signals. Every finding is tied to evidence, not opinion.
Scoring against a common five-level maturity model, triangulating across areas, weighting gaps by risk, and validating results with your stakeholders.
Executive readout, maturity index, heat map, full report, and a prioritized roadmap across three horizons: Stabilize, Build, and Operate & Optimize.
The full audit combines every assessment so the results check each other: leadership may report confidence while the workforce reports confusion, or policy may prohibit what Shadow AI shows is widespread. Each assessment can also run on its own.
The flagship engagement. All seven assessment areas, one maturity index, one executive heat map, and one risk-weighted roadmap.
Baseline your governance maturity across eight domains against a target state you choose, with a risk-weighted gap register and roadmap.
Six-dimension AI leadership capability assessment with an anonymized 360 review. Individual and cohort reports.
Map your obligations to your AI use, test the evidence, and get a readiness scorecard that reveals live legal and regulatory exposure.
Two modules: Skills (practical competence by segment) and Anxiety (trust, confidence, psychological safety). Together or either alone.
Discover unsanctioned AI use, the data being exposed, and the gap between policy and practice, with an exposure heat map.
Evaluate your governance documents for coverage, coherence, compliance, and enforceability. Produces a policy defect register.
The results are built for two audiences: executives who need the picture in minutes, and the owners who must act on it. Every score and finding is evidenced.
A headline index, current-vs-target profile across eight domains, and the three actions that matter most.
Current maturity, target, and risk weight across every domain and assessment area at a glance.
Every finding plotted by likelihood and impact, so leadership sees where risk is concentrated.
Methodology, per-area findings, evidence, scorecards, and the consolidated scoring workbook.
The discovered-AI-tool register from Shadow AI and the policy defect register.
Prioritized, sequenced, and owned. Each item maps to an implementation workstream.
A 30-minute conversation about your AI landscape, your concerns, and how an audit would be scoped for you, in full or starting with a single assessment.