SAFETY AND PRIVACY NOTES
AI front desk guardrails for Australian service businesses
Wendeal’s product direction depends on trust: transparent AI use, approved answers, human escalation, data minimisation, and reviewable handoffs.
OPERATING GUARDRAILS
The pilot must be designed around things the AI is allowed and not allowed to do.
Transparent use
Disclose AI involvement in customer-facing workflows and make human escalation easy.
Data minimisation
Collect only the details needed for the next operational step, and avoid unnecessary sensitive information.
Approved answers
Use business-approved scripts for pricing, policy, service availability, eligibility, and escalation.
Human ownership
Assign a person to review transcripts, exceptions, escalations, and workflow changes.
Blocked topics
Block diagnosis, crisis support, legal judgement, final quotes, entitlement decisions, and safety-critical advice unless a qualified human handles it.
Audit trail
Keep call summaries, review notes, and rule changes visible enough to learn from mistakes.
REFERENCE POINTS
Use official guidance as design constraints for sensitive workflows.
This page is not legal advice. It is a practical buyer-facing note that Wendeal should convert into a deployment checklist for each pilot.
Privacy: review OAIC guidance on commercially available AI products and personal information before deploying customer-data workflows.
AI safety: align pilots with Australia’s Voluntary AI Safety Standard, especially human oversight, transparency, and risk management.
Aged care: treat service navigation as a human-reviewed workflow because reforms and service pathways are changing.