Notes / Company
What should customers understand when AI helps deliver a service?
An AI label leaves practical questions unanswered. Explain what the service does, which decisions it can make, and who is responsible when a customer needs help.
By Edu Rigonato. Published Oct 2, 2026. 10 min read.
A customer sees “AI-powered” beside a service and still has practical questions. Is AI helping someone prepare the work, answering questions directly, or making decisions that affect the request? If something needs attention, who takes responsibility?
Explain the actual work, decision authority, material limitations, and route to the person responsible. Make that explanation match how the service operates. Customers should not have to guess whether an action is completed, awaiting a decision, or outside the AI's authority.
That does not mean describing every model or tool behind the service. It means giving customers enough useful information to understand the arrangement they are entering. A scheduling assistant, an AI Worker that owns an operational job, and AI used to help prepare a proposal can have very different roles. The explanation should reflect those differences.
It also does not mean promising that disclosure will increase trust. Research reports negative, null, and sometimes positive effects in different settings. The practical aim here is accurate expectations and clear responsibility. The following examples illustrate how a business might explain its service; they are not measured client results or statements about legal obligations.
What does AI actually do in this service?
Start with the job rather than the technology label. Customers need to know whether AI is involved in preparing information, communicating with them, or carrying out actions on their behalf. Those distinctions change what they can reasonably expect from the interaction.
“We use AI in our operations” is broad. “Our AI assistant answers questions about appointment availability and can book an available time” tells a customer what the interaction can accomplish. If that is how the system actually works, the second explanation gives them a practical starting point.
For work that happens behind the scenes, the explanation can be different. A service business might use AI to organize intake material before its team prepares a proposal. Describing that arrangement as an AI assistant delivering the entire service would overstate its role. Describing a Worker that completes an operational job as merely a drafting aid could understate it.
Frequently asked questions
Do we need to name the AI model in the first explanation?
The initial practical explanation should make the role and service arrangement understandable. A model name alone does not tell customers what can be completed or who is responsible. Further technical information may be relevant to a particular relationship; this article does not assess legal disclosure duties.
Should we say every AI action is reviewed by a person?
Only if that is true. A Worker can make authorized decisions and complete routine work within agreed boundaries. Explain which requests need the responsible person's judgment instead of inventing review of every step.
Is an AI badge enough?
A badge can identify AI involvement, but it does not explain the work, authority, meaningful limitations, or contact route. Pair it with operationally accurate information suited to the customer's decision.
How can we assess whether the explanation is understandable?
Ask people to describe what they believe the service can do and how they would seek help, then compare those answers with the actual arrangement. Treat that as feedback on comprehension, not proof of a trust or conversion improvement.
Meet Taurist