Foundational
Certification and risk & autonomy assessment establish shared language and a baseline.
capability blueprint
The WEKID™ capability blueprint—what to build, by maturity
This is not a WEKID product launch schedule. It is the blueprint of reference capabilities the framework defines, so partners and enterprises can plan what to build and in what order. WEKID sells certification and licenses the IP; partners and enterprises build the capabilities themselves.
maturity tiers
How the capabilities sequence
A practical order for building WEKID-based capability—foundation first, high-assurance last.
Runtime
The output governance platform and agentic guardrails put scoring and gating into execution.
Oversight
Executive dashboards and managed evaluation give leaders visibility and calibrated judgment.
High-assurance
Secure, mission-grade implementations for classified and oversight-heavy environments.
capability summaries
Reference capabilities, with build paths
Each capability below can be built by a partner to bring to market, or by an enterprise for internal use, under license.
Certification & Training Program
The credentialing and education program WEKID offers directly—technical and non-technical paths, role-based training, and scenario-based assessment.
- Practitioner and Authority certification levels
- Executive, evaluator, auditor, and inspector tracks
- Curriculum, exams, and renewal requirements
- Train-the-trainer for partners and enterprises
AI Risk & Autonomy Assessment
A time-boxed engagement that evaluates existing AI systems for epistemic failure modes, autonomy levels, governance gaps, and immediate risk exposure.
- Governance gap analysis
- Appropriate autonomy determination
- Dashboard options
- Risk registers for enterprise and smaller clients
AI Incident Prevention & Forensics
A pre- and post-incident model that identifies near-misses and performs epistemic root cause analysis after AI failures.
- Failure layer diagnosis
- Gate-miss analysis
- Corrective control recommendations
- Near-miss review support
AI Output Governance Platform (SaaS / On-Prem)
A runtime governance platform that evaluates AI outputs as they are produced or acted upon across multiple vendors and architectures.
- Parses text, decisions, and tool calls
- Scores Data, Information, Knowledge, Experience, and Wisdom
- Approves, constrains, escalates, or rejects
- Generates audit-ready evidence
- Survives model churn and architectural change
Runtime Guardrails for Agentic AI
A specialized control layer for autonomous and semi-autonomous AI systems that enforces authority limits before action is taken.
- Validates tool use justification
- Enforces retry and escalation thresholds
- Applies cost, safety, privacy, and authority gates
- Performs "Should we act?" Wisdom checks
Board & Executive AI Oversight Dashboard
A non-technical executive dashboard that surfaces trust posture, score degradation, escalation activity, and autonomy movement.
- Trust trends by system
- Declining Wisdom and Experience visibility
- Escalation frequency tracking
- Autonomy expansion or rollback
Evaluation-as-a-Service (Human + AI)
A managed evaluation model combining automated WEKID scoring with trained human evaluators for high-risk outputs.
- Supports courts, healthcare, regulatory, and enterprise decisions
- Reduces internal review burden
- Provides calibrated judgment and accountability
- SLA-backed with escalation playbooks
Secure AI (Defense & Intelligence Edition)
A high-assurance WEKID implementation pattern for classified, mission-critical, and oversight-heavy environments.
- Mandatory Wisdom gates
- Conservative autonomy defaults
- Explicit escalation chains
- Mission-aligned risk profiles
how to use this page
Plan your WEKID build path
1
Identify the right starting point
Pick the capabilities that match your current governance, evaluation, or assurance need.
2
Sequence by maturity
Use the tiers to build in a sensible order—foundation first, high-assurance later.
3
Choose partner or enterprise
Decide whether you will sell the capability (partner) or run it in-house (enterprise).
4
License and engage early
Open discussions on authorization, licensing, and design input before you build.
next step
Use the blueprint to plan what you'll build
Decide your path, take the matching license, and engage early on the capabilities that fit your priorities.