AI Policy

1. Purpose and Scope

This Artificial Intelligence (AI) Policy outlines how our consultancy uses AI tools ethically, securely, and effectively to support our business operations and client engagements.

This policy applies to all employees, contractors, and directors. It sits alongside our Privacy Policy and aligns with our obligations under the New Zealand Privacy Act 2020.

2. Core Principles

When using AI tools, we commit to the following principles:

• Human Accountability: AI assists our work; it does not replace our professional judgment. A human consultant always reviews, verifies, and approves any AI-generated deliverable or recommendation before it reaches a client.

• Data Privacy & Confidentiality: We protect our clients' and our own sensitive information at all times.

• Transparency: We are open with our clients about how and when AI is used in delivering our consultancy services.

• Cultural Respect: We respect Indigenous Data Sovereignty and handle data in line with Te Tiriti o Waitangi principles.

3. Approved AI Use Cases

Our team may use enterprise-grade, approved AI tools for internal and support tasks, including:

• Brainstorming, outlining, and drafting preliminary documents or communication.

• Summarising lengthy, non-confidential reports or research papers.

• Data analysis on aggregated or fully anonymised datasets.

• Improving internal administrative workflows and note-taking during meetings (where consent is given).

4. Prohibited Uses & Data Protection

To prevent privacy breaches and intellectual property risks, team members must not:

• Input Personal Information: Enter identifiable client information, staff details, or personal data into public or unapproved AI tools.

• Input Sensitive Business Data: Upload confidential client documents, commercially sensitive files, proprietary methodology, or security credentials into public AI tools.

• Rely Solely on AI Outputs: Use AI-generated advice, legal, financial, or strategic content without independent verification. AI is known to generate incorrect or biased outputs ("hallucinations").

• Use Public AI Model Training Options: Use AI tools where input data is saved or used to train public models. Only approved enterprise plans with strict data privacy controls may be used.

5. Client Transparency

We believe in building trust. If AI tools play a material role in generating client deliverables (such as market analysis or automated modeling), we will disclose this to the client.

6. Incident Reporting & Policy Review

If a team member suspects that sensitive or personal data has been accidentally entered into an unapproved AI tool, or if an AI output leads to an operational error, it must be reported to management immediately.

This policy is reviewed annually to keep pace with New Zealand regulatory updates, technological advancements, and industry standards.

Last updated: 1 September 2026