System summary
Name the AI feature, user group, deployment region, vendor, input data, output type, and business owner. A short summary prevents a guide result from becoming detached from the actual product behavior.
Template
AI Transparency Notice Template
Use this starter text near a chatbot, AI assistant, or generative AI feature. Adapt it to your product, data handling, and jurisdiction.
Template text:
This feature uses artificial intelligence to generate or assist with responses. AI outputs may be incomplete, inaccurate, or unsuitable for important decisions. Do not enter sensitive personal information unless the product explicitly supports that use. For questions, corrections, or human review, contact us through our support channel.
This feature uses artificial intelligence to generate or assist with responses. AI outputs may be incomplete, inaccurate, or unsuitable for important decisions. Do not enter sensitive personal information unless the product explicitly supports that use. For questions, corrections, or human review, contact us through our support channel.
Good transparency notices usually explain
- That the user is interacting with or receiving output from AI.
- What the AI feature is intended to do.
- What the AI feature should not be used for.
- Whether human review is available.
- Where users can ask questions or report problems.
Official sources
Last reviewed: 2026-07-03.
- EU AI Act Article 50: transparency obligations
- European Commission: transparency of AI-generated content
Check whether your tool needs more than a notice
Implementation example
Example review record for AI Transparency Notice Template
Use this example structure when saving a result from this guide into an internal AI compliance folder. The goal is to make the output specific enough that another teammate can understand the decision months later.
Risk and notice assumptions
List whether the system affects employment, education, credit, healthcare, biometric identification, public benefits, safety, or another high-impact area. If the page is used for transparency wording, record where the notice will appear and what users can do after seeing it.
Next review trigger
Set a clear trigger for review: model change, new vendor, new data category, new market, product launch, user complaint, policy update, or a change from internal-only use to customer-facing use.
Minimum record to keep
- Date of review and reviewer name or role.
- Generated output and any manual edits made before use.
- Source links, unresolved questions, and the reason specialist review is or is not required.
- Owner responsible for updating the record when the AI system changes.