DataCare Solutions

DataCare Solutions Pte. Ltd.
UEN: 202530943R
Website: www.datacaresolutions.biz
Email: info@datacaresolutions.biz

AI System Inventory

Comprehensive Register of AI Systems and Tools

About This Inventory

This AI System Inventory template provides a structured register for documenting all artificial intelligence systems, machine learning models, and AI-powered tools deployed within your organisation. Maintaining a comprehensive inventory is a foundational requirement for AI governance, risk management, and regulatory compliance.

How to Use This Inventory

Record every AI system, tool, or model currently in use or under evaluation

Assign a risk level to each system based on the definitions provided on the following page

Identify a System Owner responsible for governance and oversight of each entry

Document all data inputs and outputs to understand data flows and exposure

Set a Next Review Date for each system (recommended: quarterly for High/Critical, biannually for Low/Medium)

Update this inventory whenever a new AI system is procured, deployed, or decommissioned

Important: This inventory should be reviewed at least quarterly and kept as a living document. It forms the basis for AI risk assessments, audit trails, and compliance reporting. Ensure all stakeholders have access to the latest version.

AI System Register

#	AI System / Tool Name	Vendor / Provider	Business Purpose	Data Inputs	Data Outputs	Risk Level	System Owner	Deployment Date	Next Review Date
1	ChatGPT Enterprise	OpenAI	Customer support ticket triage and suggested responses	Customer enquiry text, ticket metadata, account details	Suggested replies, ticket category, priority score	Medium	Head of Customer Support	15 Jan 2025	15 Apr 2025
2	HubSpot AI (Lead Scoring)	HubSpot	Automated lead scoring and sales pipeline prioritisation	Contact details, website behaviour, email engagement, company data	Lead score, conversion probability, recommended actions	Low	Marketing Director	01 Mar 2025	01 Sep 2025
3	Internal ML Model (Fraud Detection)	In-house / Data Science Team	Real-time transaction fraud detection and flagging	Transaction amount, merchant data, user behaviour patterns, device fingerprint	Fraud risk score, block/allow decision, alert notifications	Critical	Chief Technology Officer	10 Nov 2024	10 Feb 2025
4
5
6
7
8

Risk Level Definitions

                    Low Risk

                    AI system processes only non-personal or publicly available data. No automated decision-making that materially affects individuals. Minimal regulatory exposure. Examples: internal content drafting tools, code auto-completion, spell-check assistants.
                
                    Medium Risk

                    AI system processes personal data or influences business decisions. Outputs are reviewed by a human before action is taken. Moderate regulatory exposure. Examples: lead scoring, customer segmentation, AI-assisted recruitment screening with human review.
                
                    High Risk

                    AI system processes sensitive personal data or makes recommendations that significantly affect individuals. Limited human oversight in the decision loop. High regulatory exposure. Examples: credit scoring, automated HR decisions, medical diagnosis support.
                
                    Critical Risk

                    AI system makes fully automated decisions with material impact on individuals' rights, finances, or safety. Processes large volumes of sensitive data. Maximum regulatory exposure. Examples: real-time fraud blocking, autonomous surveillance, biometric identification systems.

Review Schedule

A regular review cycle ensures your AI inventory remains accurate and that risk assessments reflect current usage. The following quarterly review checklist is recommended:

Quarterly Review Checklist

Verify all AI systems in the inventory are still active and accurately described

Identify and add any new AI systems deployed since the last review

Remove or archive decommissioned AI systems

Re-assess risk levels based on any changes in data inputs, outputs, or scope

Confirm each System Owner is still the correct responsible party

Review vendor contracts and data processing agreements for AI tools

Check for regulatory updates that may affect risk classification

Document review findings and update the Next Review Date for each entry

Escalate any newly identified High or Critical risk systems to senior management

File a copy of the updated inventory with the Data Protection Officer

Keeping this inventory current

AI governance is not something DataCare Solutions sells as a separate service.
Our free AI governance guide explains how to keep this register useful,
and staff training on safe AI use is covered in AI Training.

DataCare Solutions Pte. Ltd.

Website: www.datacaresolutions.biz
Email: info@datacaresolutions.biz

Document Version: 1.0 | Last Updated: February 2026
Prepared by: DataCare Solutions Pte. Ltd.

© 2026 DataCare Solutions Pte. Ltd. All rights reserved.
This template is for informational purposes only and does not constitute legal advice.