“If I used Claude to write a first draft, do I need to say so?”
“If ChatGPT helped me organize my thoughts, but I did the writing myself, do I need to explain that?”
“But if I disclose that I used AI, won’t readers think less of my work, even if it’s awesome?”
The question of whether or not to tell readers that you used AI in your writing is a deceptively simple one. Reasonable people disagree about how much disclosure is necessary. Some argue that writers should always be open about AI use, no matter how trivial. Others think that readers should judge the quality of the work itself, not how it was produced. There’s no real standard here…at least not yet.
This article offers a framework for thinking through your own feelings about disclosure, and whether and when to acknowledge AI use in your writing.
Pros and cons of disclosing AI use
At first blush, it might seem that complete transparency all the time is the way to go. But disclosing the use of AI in creating content also has downsides. As with many questions about AI, the decision isn’t always straightforward. Consider:
Reasons to disclose
- Transparency: Being open about AI use can demonstrate honesty and build trust with your readers.
- Personal comfort: You won’t be anxious about whether you’re hiding or withholding something.
- Policy compliance: Disclosure may be required by your employer, client, or publisher.
Reasons not to disclose
- Reputational risk: Some people may view AI use negatively and assume you took shortcuts or “cheated.”
- Reader bias: Disclosure may lead some readers to distrust content they would otherwise find useful and credible.
- Changing expectations: As AI becomes more common, some level of AI assistance may become assumed (similar to the use of tools like spelling and grammar checkers).
Deciding whether (and what) to disclose
So, what considerations matter when making this decision? These questions can help you think through the details:
- Did AI make a substantial contribution to the final content?
- Would readers feel surprised or misled if they learned that AI was used?
- Is authorship, expertise, or personal experience central to the piece?
- Did AI contribute more than just writing assistance?
- Could errors in this content have significant consequences?
- Does your employer, client, publisher, or profession require disclosure?
In general, the more “yes” answers you have, the stronger the case for disclosure. But these questions aren’t intended to produce a definitive answer. I imagine you can think of exceptions and edge cases that blur the line.
It’s also worth remembering that disclosure is not an all-or-nothing choice. The right approach may depend on how AI was used, who the audience is, the purpose of the content, and the expectations of your profession or organization.
Like everything with AI, your human judgment remains essential.
How to acknowledge AI use in content creation
You’ve decided to disclose that you used AI. Now what? It depends on how much information you want to share and why.
Options include:
1. Simple acknowledgment
A brief, straightforward statement that AI was used.
Examples:
- “This article was developed with the assistance of generative AI tools.”
- “AI was used to support the development of this content.”
2. Purpose-based disclosure
Provides more useful context by explaining what AI was used for.
Examples:
- “AI was used to help generate ideas and organize the initial outline.”
- “AI was used to summarize source materials; all content was reviewed by the author.”
3. Contribution-based disclosure
Describes the extent of AI involvement and distinguishes AI’s role from the author’s role.
Examples:
- “The initial draft was generated with AI and substantially revised by the author.”
- “AI helped draft selected sections; all final editorial decisions were made by the author.”
4. Process disclosure
Describes the workflow in more detail for full transparency.
Example:
“The author used ChatGPT to generate alternative structures, identify content gaps, and suggest revisions. All factual claims were independently verified, and the final content was written and approved by the author.”
5. Conversational disclosure
This more informal approach often matches audience expectations on platforms like LinkedIn and on personal blogs.
Examples:
- “I used AI to help brainstorm this post.”
- “AI helped me work through some early drafts, but the opinions here are my own.”
6. Embedded disclosure
Instead of a formal, separate note, mention AI use within the text itself.
Examples:
- “While experimenting with ChatGPT, I noticed…”
- “I used Claude to help analyze the responses…”
Looking Ahead
I suspect that norms around AI disclosure will settle down pretty quickly. Think about ghostwriting—there used to be a lot of debate about whether a ghostwriter should be acknowledged. These days there’s a much clearer distinction between when ghostwriting is generally considered okay or is even expected (e.g., celebrity memoirs, political speeches) and where it’s unethical (e.g., academic papers).
If I had to guess, I’d say that AI disclosure standards will primarily center on these two factors:
- What role AI played—was it just a tool, or a substitute for the author?
- Whether AI use is likely to affect reader trust or interpretation.
And some fields will have stricter professional disclosure policies than others. I’d imagine that professions like law or medicine, which tend to have higher stakes and require more precision, will have a higher bar than marketing or blogging.
In the meantime, be intentional about whether and when you acknowledge AI use in your writing. As a reader, what kind of disclosure would you want to see? As always, you can’t go too far wrong by writing with your reader in mind.
Disclosure: I’ll go first: I used ChatGPT to help brainstorm, organize, and edit this post, and to create wording for specific examples. I substantively wrote and edited each section. Views and opinions are my own.