Responsible AI

Governance belongs in the process, not in a policy binder

BEE Sigma’s implementation method designs in purpose, data flow, permissions, sources, human review, logging, exceptions and vendor boundaries. Specific controls depend on project risk, technology and contract scope.

01

Purpose & data minimization

Scope the business purpose first, then assess what data can be reduced, de-identified or simply not collected.

02

Sources & accuracy

Key answers keep their sources, versions and test evidence wherever possible; when evidence is thin, design refusal or hand-off paths.

03

Human review & accountability

For high-impact decisions, name an accountable approver and keep the responsibility boundary in the design.

04

Permissions & isolation

Assess isolation by role, client and task — with least privilege as the design target.

05

Logs, exceptions & rollback

Keep the operating evidence the risk and contract require; design exception handling, human fallback and rollback paths.

06

Ongoing review

Re-assess when models, knowledge, vendors or purpose change; one acceptance test is never a permanent guarantee.

Data Flow · Map the critical path

Make the important steps explainable and reviewable

Within project scope, record the key input, processing, review and retention steps. Specific vendors, regions, retention periods and training usage must be confirmed per project and contract.

01Inputsource, purpose, necessity
02Processingmodels, rules, vendors
03Reviewowner, exceptions, refusals
04Retain / Deleteduration, logs, exit
Deployment · More than the word “private”

Choose by data, control and operational responsibility

A

Customer / controlled cloud

Run in a cloud and region the client approves, with identity, logging, keys and vendor responsibility made explicit.

B

On-premise or offline

Use the word “offline” only when models, hardware, updates, operations and feature limits are all explicit.

C

Managed service

Make data flow, sub-processors, retention, exit and service responsibility explicit; convenience never replaces scrutiny.

NZ · NEW ZEALAND CONTEXT

A Privacy Impact Assessment is a living design document

New Zealand’s Privacy Commissioner recommends conducting — and continuously updating — a PIA before using AI on personal information. When the process, fields, vendors or purpose change, so should the assessment.

Hand your first workflow to the agents

Decide where it’s worth doing before deciding which AI. One free scan plus one adoption assessment — governance designed in from day one.