If AI is involved, you’re told.
Plainly Disclosed is the guardrail that keeps trust simple: customers and staff should never have to guess whether they are dealing with an AI system, reading AI-generated content, or affected by an AI-influenced decision.
What plain disclosure actually looks like
Told, not left guessing
Customers are told when they are talking to an AI agent, in plain language, not buried in terms and conditions.
AI-generated content is labelled
Content generated by AI is labelled as such wherever it matters to the person reading it.
Decisions are disclosed
Where AI has influenced a decision about someone, that is disclosed to the person it affects, plainly.
Built in, not left to memory
Disclosure is written into the system itself, so it happens every time, not only when a staff member remembers to mention it.
Being upfront builds trust. Being caught hiding it destroys it
Most customers are comfortable dealing with an AI agent once they know that is what is happening. What actually erodes trust is finding out afterwards that they were not told. This guardrail keeps that choice off the table.
Grounded in
- Australia’s Voluntary AI Safety Standard, guardrail 6 (inform end-users)
- EU AI Act transparency obligations for limited-risk systems, e.g. chatbot disclosure
The rest of the framework
All ten guardrails run through every engagement. Here are the other nine.
Frequently asked questions
Won’t disclosure make customers trust the system less?
Usually the opposite. Being upfront about it is what makes people comfortable using it. Hiding it is what erodes trust when they find out anyway.
Do we need to disclose AI use for internal, back-office tools too?
Less critical for pure back-office tools, but staff should still know they are working with an AI system, so they know how and when to escalate.
Find out where your governance actually stands. Governance is one of six things the free audit scores.
Get the free audit