An AI system is an AI-powered solution that is developed, deployed, and managed under a formal governance framework. This framework ensures that the system operates in a responsible, compliant, and risk-aware manner throughout its life cycle.

An AI system functions within an organization’s digital or operational environment to support business activities. It enables automation, enhances data-driven decision-making, and improves overall process efficiency. The system includes components such as machine learning models, natural language processing engines, or computer vision tools.

These components are often supported by datasets, algorithms, and computing infrastructure. Together, they enable the system to analyze data, recognize patterns, and generate outcomes with minimal human intervention. Each AI system is subject to a structured evaluation process. This process assesses potential risks related to functionality, security, and societal impact. It also verifies compliance with applicable regulations, industry standards, and internal governance policies.

Governance frameworks are applied to manage the AI system throughout its life cycle. They ensure that the system operates securely and delivers predictable, explainable, and trustworthy results. Ethical principles such as fairness, accountability, and transparency are enforced through defined controls and oversight mechanisms. Regular monitoring supports responsible AI use and reduces the risk of unintended consequences.

The following image shows the overview page of an AI system.

Aggregated risk score consolidates individual risks such as bias, drift, and security, to inform departmental or enterprise-level AI risk profiles, enabling higher-level visibility and oversight. For example, several customer-facing AI models exhibiting signs of bias can lead to organizational risks. Aggregated risk score enables the AI Risk and Compliance team to obtain a consolidated view of AI risks across multiple models, teams, and business units, moving beyond fragmented risk assessments.

The following image shows the aggregated risk score section: