Our Mission
To make AI skill measurable, developable, and continuous. Not a certificate you earn once.
AI Vitals is a continuous AI skill development platform. We measure literacy and fluency across six validated dimensions, then keep your score current as you learn, practice, and reflect. Every lesson you complete, every reflection you write, every practice session you run updates your skill picture — no retake required.
Why a 6-dimension framework?
Most AI assessments treat AI literacy as binary: you either “know AI” or you don't. AI skill doesn't work that way. You can have solid mental models for how AI works (L1) and still fail to calibrate trust in its outputs (F2). A single score misses the specific gaps that matter for workplace readiness.
Our framework separates Literacy (conceptual understanding: L1, L2, L3) from Fluency (practical skill: F1, F2, F3), giving organizations a precise, actionable picture of where to invest — at the individual, team, and org level.
A Living Score, Not a Snapshot
Traditional assessments give you a number that decays the moment the window closes. AI Vitals tracks a continuous stream of domain signals, tagged to specific dimensions, weighted by engagement quality, and accumulated over time.
When you complete a lesson, write a reflection, run a practice session, or have a coaching conversation, those touchpoints update your skill picture automatically. Your score evolves with you — and managers can see projected team readiness based on current learning velocity, not just the last assessment date.
Built with AI, for AI learning
AI Vitals is built end-to-end with Anthropic's Claude. Thirteen AI-powered features share a common memory of who you are and where you stand — from a coaching debrief after your first assessment, to an executive ROI report after six months of team-wide learning.
The best way to teach people to work with AI is to show them. Every learner gets a coach, an advisor, and a practice partner that remember their scores, their reflections, and their growth arc.
Designed for Every Organization
Individual professionals and 500-person enterprise teams use the same platform. Start with a free assessment; add team dashboards, manager reporting, spend controls, and executive forecasting as you grow.
We offer sponsored access for education institutions, training providers, and workforce development programs — because the AI skills gap isn't evenly distributed.
Who We Built This For
Individual Contributors
Get a precise picture of your AI skills, a personalized action plan, a practice partner for hands-on improvement, and an advisor that learns your workflows and spots AI opportunities you're missing.
L&D Teams & Managers
See your team's aggregate strengths and gaps, generate AI-written synthesis memos, and track learning velocity over time. Replace manual reporting with data that updates itself.
Executives & HR Leaders
Get a 3-, 6-, and 12-month AI readiness forecast, an executive ROI narrative, and full platform oversight — spend limits, audit logs, SSO, and cross-org user management.
The Assessment Framework
Six dimensions. Each tracked continuously by learning signals from lessons, practice, and conversations.
| Code | Dimension | What It Measures | Type | Signal Sources |
|---|---|---|---|---|
| L1 | AI Foundations & Mental Models | Functional understanding of what AI can and cannot do, how outputs are generated, and calibrated expectations across tool types | Literacy | Lessons, assessments |
| L2 | Ethics & Responsible AI | Bias recognition, data privacy, fairness, societal impact, governance, and organizational accountability | Literacy | Reflections, advisor conversations, assessments |
| L3 | Strategic & Tactical AI Judgment | Knowing when, where, and whether to apply AI at both the task level and the organizational level; task decomposition, AI-suitability, cost-benefit reasoning | Literacy | Advisor conversations, reflections, assessments |
| F1 | AI Interaction & Iteration | Structuring effective human-AI interactions — initial prompt design, iterative refinement, conversation management, and knowing when to restart vs. continue | Fluency | Practice attempts, lessons, assessments |
| F2 | Critical Evaluation & Calibrated Trust | Assessing AI output quality, detecting hallucinations, verifying claims, resisting the artifact paradox, and calibrating trust proportionally to stakes | Fluency | Practice attempts, reflections, assessments |
| F3 | Workflow Integration & Orchestration | Embedding AI into repeatable work processes, building reusable systems, designing human-AI handoffs, and orchestrating multi-tool and multi-agent workflows | Fluency | Advisor conversations, lessons, reflections |
How AI Vitals Learns About You
Every interaction builds a richer picture of your skills, habits, and growth trajectory — here's the mechanism.
Domain Signals
Every learning touchpoint emits a signal tagged to one or more dimensions with a quality weight: a lesson completed, a reflection submitted, a practice session run. Those signals accumulate into a continuously updated skill picture.
Advisor Memory
Your AI Work Advisor maintains a structured memory of your role, workflows, constraints, and demonstrated capabilities. When you describe how you used AI in a real task, that becomes a durable memory your advisor and coach remember.
Lesson Reflections
After each micro-lesson, a 30-second reflection prompt asks: "What will you do differently?" Your answer is extracted into a capability memory — concrete evidence of applied learning, not just completion.
Assessment Anchors
Formal assessment retakes anchor your signal history to verified scores. The platform uses these anchors to calibrate the weight of ongoing signals — giving you an honest trajectory, not an inflated one.
Built with Anthropic Claude
All AI features in AI Vitals are powered by Anthropic's Claude — chosen for its safety, instruction-following, and long-context reasoning capabilities.
Responsible AI Commitment
AI Vitals is used in employment contexts. Those come with regulatory and ethical obligations, and we take them seriously.
EU AI Act — Article 4
AI Vitals is classified as a high-risk AI system under Annex III, Category 4 of the EU AI Act (AI systems used for employment, worker management, and access to self-employment). We comply with Article 4 obligations including:
- Human oversight of all employment-affecting decisions
- Right to explanation for every assessment result
- Annual bias audit against demographic dimensions
- Technical documentation and conformity declaration available on request
NYC Local Law 144 & Bias Auditing
Employers using AI Vitals scores in employment decisions in New York City must comply with Local Law 144, which requires annual independent bias audits of automated employment decision tools (AEDTs). AI Vitals provides:
- Built-in bias audit export (admin panel → Bias Audit tab)
- 4/5ths adverse impact ratio calculation per demographic group
- k-anonymity protections (groups < 5 suppressed)
- Audit trail with immutable log entries
Psychometric Transparency
Our assessment is a self-report learning diagnostic — not a validated psychometric instrument for high-stakes employment decisions. Every score is displayed with a ±9 point standard error of measurement and a plain-language disclaimer. We are conducting an IRT calibration study and will update our documentation as results become available.
Data Governance
We process assessment data under GDPR Article 6(1)(b)(contract performance). We do not use your data to train Anthropic models — we are an API customer subject to Anthropic's Data Processing Addendum. Assessment scores are retained for 3 years; coaching conversations for 90 days.
Ready to measure your team's AI readiness?
Start free in 20 minutes. If you're rolling it out across a team, request a demo.