Should a Business Disclose When Website Content Was Created with AI?
Decide when AI disclosure helps customers understand website copy, images and evidence, with practical wording and checks for UK business marketing.
Part of our guide: How to Use AI for Website Content Without Losing Trust or Quality
A business should explain its use of AI when that information affects how a customer understands or relies on the content. The clearest cases involve synthetic people, illustrative images that could look like real work, and automated conversations.
Editing a paragraph with AI does not create the same expectation as inventing a customer testimonial. Start with what readers are likely to believe, then make the source and status of the material clear.
Ask what the reader might reasonably infer
A photograph beside a case study usually suggests that the business carried out the work shown. A quotation with a name suggests a real person said it. A first-person account suggests the writer experienced the event.
If AI helped create any of these, check whether the presentation remains accurate. Labelling a fabricated review as AI-generated does not turn it into customer evidence. Replace it with a real approved review or an explicitly fictional example.
For decorative illustrations, a general explanation of your editorial process may be enough. For a simulated result, the explanation needs to sit beside the image where the reader makes that judgement.
Thinking about your own rebuild?
Tell us about your business, your current website and what you would like to improve. We will help you work out the next step.
What does UK advertising guidance say?
The ASA’s guidance on AI disclosure focuses on whether an audience would be misled and whether a disclosure clarifies the message. Its central point is that existing advertising rules apply regardless of how an advertisement was made.
A label cannot repair an underlying false impression. A generated before-and-after image that exaggerates a product’s effect remains a problem even if a footnote mentions AI.
Check the rules for your sector, distribution platform and target countries before using synthetic endorsements or realistic depictions. Different contexts can require different treatment; a website policy is not a universal permission slip.
Describe the actual contribution
Use specific wording that reflects the process. Possible examples include:
- “Illustration created with AI and reviewed by our team. It does not depict a completed customer project.”
- “This fictional example explains the process; the figures are illustrative.”
- “Our team uses AI to help organise drafts. Named staff review factual claims and approve the published content.”
- “You are speaking to an automated assistant. You can contact our team here.”
These are examples to adapt, not statements to copy unless they are true. Do not claim expert review if nobody with the relevant knowledge checked the work.
Put the explanation where it matters
An image caption is useful when the image’s status could change a buying decision. An editorial policy is useful for explaining a recurring process. A chat introduction is useful before somebody starts sharing information with an automated service.
Avoid burying a material explanation in terms and conditions. Equally, repeated labels on every minor spelling correction may distract from the facts a customer actually needs.
Ask someone unfamiliar with the page what they think the image, quotation or account represents. If their interpretation differs from reality, revise the presentation.
Disclosure does not replace review
Accuracy, permissions and responsibility still need checking. Verify claims, secure rights to source material and decide who can approve publication.
The separate guides to fact-checking AI content and ownership of AI-created copy and images cover those decisions. A disclosure describes the production process; it does not provide evidence for a claim or a licence for an image.
Build a small editorial rule
A practical policy can separate routine drafting assistance from material synthetic content. Require an explicit review for testimonials, case studies, staff portraits, demonstrations and product results. Record the decision beside the asset so future editors do not accidentally remove its context.
For example, a landscaping business could use a paper illustration to explain drainage principles, but should label a generated garden visual as a concept if visitors might mistake it for a completed installation.
Should every AI-assisted article carry a label?
There is no useful one-size answer for all content and jurisdictions. Consider the expectations created by the page, relevant rules and the significance of the assistance. Be especially clear where the page claims personal experience or presents something as documentary evidence.
Google’s guidance on generative AI content also encourages useful context about how content was made. That is an editorial consideration, not a promise of improved rankings.
Make the explanation proportionate, visible and accurate. Our guide to using AI without losing trust explains the wider workflow, and you can contact AfterWorks to discuss your website content.
