Plain-language definitions.

Any artificial intelligence model, software application, or integrated SaaS utility tracked within the organization's inventory, encompassing both formally registered systems and those flagged in the discovery queue for review.

A structured, signed summary detailing an AI system's current governance status, including its risk profile, active policies, verified control states, compiled evidence, and historical approval records.

An append-only, tamper-evident log that records all platform activities, documenting who performed each action, when it occurred, which record was affected, and the underlying operational justification.

An outstanding regulatory or internal policy requirement flagged when a framework is applied to an AI system, requiring a resolved control, verified evidence, or an approved temporary exception.

A secure, versioned package containing updated regulation texts, compliance templates, control frameworks, and discovery definitions, delivered automatically to cloud environments or imported manually into air-gapped systems.

A technical, procedural, or contractual safeguard implemented to mitigate a specific risk or satisfy a regulatory obligation, rated by effectiveness and backed by verified compliance evidence.

A legally binding contract between a data controller and a processor defining the boundaries of personal data processing, security baselines, sub-processor disclosures, and breach notification windows.

An organization utilizing an AI system under its own authority, carrying distinct compliance obligations under frameworks like the EU AI Act that differ from those of system developers or providers.

Data Protection Impact Assessment. A structured analysis required when processing is likely to result in high risk to individuals' rights. Mandatory under GDPR for many AI deployments involving personal data.

A verified artifact, such as a configuration log, screenshot, bias test report, or signed policy document, stored securely in the repository to prove that a compliance control is active and functional.

A strictly time-limited waiver for an open compliance gap that requires a documented business justification, formal approval, and a firm expiration date, after which the gap automatically reopens.

A real-time metric indicating the ratio of verified, human-approved compliance controls to total active regulatory obligations across all registered AI systems.

A classification defined by regulatory frameworks (such as Annex III of the EU AI Act) for AI deployments in sensitive domains, such as employment, education, credit scoring, or critical infrastructure, that require strict human oversight, detailed documentation, and continuous monitoring.

The designated individual held accountable for a policy's lifecycle, including drafting, periodic review, exception approvals, and eventual retirement.

An entity that develops or commissions an AI system and markets or deploys it under its own brand name, carrying the primary legal burdens under global AI regulations.

A platform component that monitors regulatory updates within imported Content Bundles, flagging affected systems, controls, and policies that require compliance review.

The mandatory frequency at which a system's context, risk assessment, or active controls must be re-evaluated, with overdue reviews automatically triggering dashboard alerts.

An impact-based classification assigned to an AI system based on structured context reviews, dictating the depth of necessary controls, required evidence, and review frequency.

Unregistered or unmanaged AI tools, models, or third-party integrations currently used across an organization without formal assessment, authorization, or tracking.

A workflow protocol requiring that the user who submits a risk assessment, governance plan, or evidence package cannot be the individual who reviews and approves it, preventing conflicts of interest.

A compliance document where every revision automatically generates a new, distinct version, permanently preserving past historical iterations alongside timestamps and approver details.