A precise language for running an AI estate.
Cybatar definitions for the concepts institutions need to govern, evidence and economically manage consequential AI.
Enterprise AI Management
Enterprise AI management is the institutional discipline and operating layer for inventorying, governing, evidencing, operating and economically managing AI across an organisation.
Read definition →AI Estate
An AI estate is the complete set of AI systems an organisation builds, buys, accesses or operates, together with their owners, suppliers, dependencies, risks, costs, controls and lifecycle status.
Read definition →AI Control Tower
An AI control tower is an enterprise management layer that gives leaders consolidated visibility and decision control across the organisation’s AI portfolio, risks, economics and operating status.
Read definition →AI Passport
An AI Passport is a governed system dossier that brings together an AI system’s identity, purpose, owner, model/provider dependencies, lifecycle, risks, controls, evidence, economics and decision history.
Read definition →AI Governance
AI governance is the system of policies, roles, decision rights, controls, reviews and evidence used to direct and oversee the development, procurement, deployment and use of artificial intelligence.
Read definition →Runtime AI Governance
Runtime AI governance is the observation and enforcement of AI policies, permissions, moderation, evidence and escalation while AI systems are actually operating.
Read definition →AI Evidence
AI evidence is the verifiable record of material AI facts and activity, including decisions, sources, model/provider context, tool calls, calculations, moderation, human review and integrity information.
Read definition →AI Assurance
AI assurance is the structured evaluation and evidence process used to determine whether an AI system, control or supplier meets defined expectations for risk, performance, governance and reliability.
Read definition →Reality Gap
The Reality Gap is the difference between demonstrated technical capability and proven institutional value in the Cybatar AI Reality Curve methodology.
Read definition →AI Token Economics
AI token economics is the measurement of token and model consumption in relation to workload cost, business outcomes, efficiency and validated value.
Read definition →Shadow AI
Shadow AI is AI usage that occurs outside the institution’s approved inventory, procurement, security or governance processes.
Read definition →Model Governance
Model governance is the lifecycle control of models through inventory, ownership, approval, validation, versioning, monitoring, change management, limitations and retirement.
Read definition →AI Procurement
AI procurement is the governed sourcing and contracting of AI models, applications, infrastructure and suppliers with risk, assurance, economics, usage and dependency considered together.
Read definition →No matching glossary terms.
Need these concepts applied to your institution?
Cybatar connects AI inventory, policy, accountable decisions, evidence, assurance and economics so governance can operate continuously rather than live in disconnected documents.