When to tell people you used AI (and how)

The Question That Keeps Coming Up

As AI becomes woven into everyday work, one question surfaces again and again: Do I need to tell people I used AI for this?

The honest answer is that it depends on the context. But the honest default — and the one that will serve you best in the long run — is openness. Let us explore why, and how to put it into practice.

When to Disclose AI Involvement

Different contexts carry different expectations, but here are the main areas where disclosure matters:

Professional contexts. If you are producing reports, proposals, analyses, or communications for your employer or clients, consider whether the people receiving your work would want to know that AI played a role. In many cases, the answer is yes — not because they would think less of the work, but because they deserve to understand the process behind it. This is especially true when the work involves judgement calls, recommendations, or analysis that others will rely on for decision-making.

Academic work. Educational institutions are rapidly developing their own policies on AI use, and these vary considerably. Some encourage AI as a research and drafting tool; others restrict it heavily. Regardless of the specific policy, the principle is clear: if you are submitting work for assessment, the assessor needs to know what is yours and what was assisted. Failing to disclose is not just a policy violation — it undermines the entire purpose of learning, which is to develop your capabilities.

Creative collaboration. If you are producing creative work — writing, design, music, art — the expectations around AI disclosure are still forming. But audiences and collaborators increasingly want to know. A piece of writing that was substantially drafted by AI is a fundamentally different creative product from one that was written by a human with AI assisting at the margins. Being upfront about AI's role is not a weakness; it is a sign of confidence in your own contribution.

Client deliverables. When you are being paid for your expertise, clients are paying for your judgement, your skill, and your accountability. If AI played a significant role in producing a deliverable, your client has a right to know. This does not diminish the value of what you have produced — but it does allow the client to make informed decisions about how much trust to place in the output and what additional review might be warranted.

The Spectrum of Disclosure

Transparency is not all-or-nothing. Think of it as a spectrum, and choose the level that is appropriate for the situation:

"AI-assisted" is the lightest touch. It signals that you used AI tools somewhere in your process, without going into detail. This might be appropriate for routine internal communications, brainstorming outputs, or early-stage drafts that underwent significant human revision.

"AI-generated with human editing" provides more detail. It tells the reader that AI produced a substantial portion of the content, and that you then reviewed, refined, and took responsibility for it. This is suitable for many professional documents, reports, and written deliverables.

Full process transparency is the most detailed approach. Here, you describe which tools you used, what tasks AI performed, what you contributed, and how you verified the output. This level of disclosure is appropriate for academic work, published research, high-stakes client deliverables, and any situation where the integrity of the process is as important as the quality of the output.

Building Trust Through Honesty

Here is something that might surprise you: in most professional contexts, being transparent about AI use actually increases trust rather than diminishing it.

Think about it from the other person's perspective. If a colleague tells you, "I used AI to draft this report and then spent two hours reviewing, fact-checking, and refining it," you know several things. You know they are efficient. You know they are thorough. You know they are honest. And you know they take responsibility for their work.

Now imagine discovering after the fact that someone used AI without telling you. Even if the work is excellent, the lack of transparency creates doubt. What else might they not be telling you? Can you trust their process? The quality of the work has not changed, but the relationship has.

Honesty about your process is an investment in your professional reputation. It pays dividends over time.

What to Include in a Transparency Statement

When you do disclose AI involvement, aim to cover these elements:

• Which tools you used. Name the specific AI tools. "I used Claude to assist with drafting" is more useful than "I used AI."
• What role AI played. Be specific about the tasks AI performed. Did it generate a first draft? Summarise source material? Check for errors? Suggest structure? The more precise you are, the more useful your disclosure becomes.
• What human review occurred. Describe your own contribution. Did you fact-check the output? Rewrite sections? Add your own analysis? Verify against primary sources? This is where you demonstrate the value you added beyond simply prompting an AI.
• Your affirmation of responsibility. Make it clear that you stand behind the final output. You reviewed it, you are satisfied with its quality, and you take responsibility for any errors or shortcomings.

A well-crafted transparency statement does not apologise for using AI. It demonstrates professionalism, integrity, and confidence in your process.