Earlier this week, we were discussing a fairly ordinary AI use case: using AI to help review and improve professional materials. There was agreement that technology could make process faster and better. Then, a suggestion: AI could stay in the background. The person receiving it does not need to know AI’s involvement.
This moment caught my attention.
Not because AI is controversial. Because hiding its role seemed to create comfort. It may make me wonder:
When should the person receiving professional work know that AI materially contributed to it? Even when disclosure is not required.
This is not about disclosing every tool
Professionals already use dozens of invisible tools. We do not disclose every search, spreadsheet, grammar checker, or software platform used along the way.
But AI is different in one important respect: It increasingly participates in work we historically associated with human expertise.
It can analyze. Critique. Synthesize. Recommend.
And this creates two very different transparency questions.
1. What happened to my data?
If someone gives you a proposal, capability statement, resume, engineering document, or other business information for review, and you run it through an AI system, have they agreed to that use?
The document may be public. The AI environment may be secure. The risk may be negligible.
But it is still someone else’s information being processed in a way they may not expect.
That is a data-transparency question.
2. Who actually did the thinking?
There is also a meaningful difference between:
- AI helps an expert think.
- AI performs much of the analysis, while the expert presents the result as their own.
The line between the two will not always be obvious.
And I don’t think every use of AI needs a disclaimer.
But there is a useful test:
Would knowing that AI materially contributed change how the recipient interprets the expertise, service, or advice they are receiving?
If the answer is yes, transparency probably matters.
The uncomfortable question
Organizations are increasingly comfortable using AI to make professionals faster and more productive. They are often less comfortable making AI’s contribution visible.
Why?
Sometimes the answer is trust, privacy, or client expectations. But sometimes it may be something more uncomfortable:
Making AI visible forces us to define what the human is actually adding.
Perhaps that is exactly the conversation we need.
My perspective
In my own work, I am a proponent of being explicit about where AI contributes. Because I believe transparency becomes more important as AI takes on a greater share of the work.
We need to become much more confident about articulating the human value that sits on top of an increasingly capable AI baseline.
If AI can research, synthesize, critique, and generate a credible first answer, those capabilities are no longer enough, by themselves, to define expertise.
Our value has to come from what I add beyond that baseline: context, experience, judgment, challenging the output, seeing what the model misses, and taking accountability for the recommendation.
Acknowledging AI does not diminish value. Hiding AI’s contribution does.
The durable human value proposition is not: “I did everything myself”.
It is: “I know what to do with what AI can now do.”