Trust & AI Act
Your cultural authority
stays yours
MuseMe is built for public institutions and European rules. Your sources, your curators and your visitors stay in control.
Get the white paperSix commitments, on every story
What we promise every institution we work with.
Your sources only
Stories are built from your validated content, not the open web.
Human review
Curators approve every story before it reaches a visitor.
AI disclosure
Visitors always know when content is AI-assisted, as Article 50 of the EU AI Act requires.
GDPR
Minimal visitor data, collected with consent and never sold.
EU data storage
Everything is stored in Europe: your sources, your stories and your visitor data.
Accessibility
Audio and text in the visitor's language, with described versions for blind and partially sighted visitors.
Our AI principles
Five rules we never break
How we decide what MuseMe will and will not do. Written for museums, heritage sites and the public bodies that fund them.
- 01
Sources before style
A story is only as good as what it is built on. If a claim has no source, it does not go live.
In practiceEvery story shows the sources it was drafted from. - 02
Curators decide
MuseMe drafts. Your team approves, edits or rejects. The final word is always human.
In practiceNothing reaches a visitor without your approval. - 03
Say when it is AI
Every AI-assisted story is labelled for visitors, in plain language.
In practiceA clear AI label on every story, in the visitor's language. - 04
Collect less
We only keep the visitor data needed to personalise the visit. Nothing is sold.
In practiceConsent first. Hosted in Europe. Never sold. - 05
Built for every visitor
Languages, reading levels and described audio are part of the product, not an add-on.
In practiceDescribed and multilingual versions from day one. - For procurement teamsGet the white paper
Put these principles in your tender
Trust by Design turns them into questions you can ask any AI vendor.
What the EU AI Act means for your visit
A short guide for museum teams. The white paper has the full detail.
Transparency
Visitors must know when they are hearing AI-generated or AI-assisted content.
Every story is labelled as AI-assisted by default, in the visitor's language.
Human oversight
People, not systems, stay in control of what the public sees.
Every story passes curator review. You can also build your own workflow around your current process and approval stages.
Key dates
- In force
AI literacy duty (Article 4) applies.
- In force
Article 50 transparency obligations apply.
- Next
AI-generated content must be marked, including from tools already in use.
- Coming
High-risk obligations (Annex III) apply.
Source: AI Act (Regulation 2024/1689) and Digital Omnibus on AI (Regulation 2026/1744). Not legal advice.
Questions we hear
What directors, curators and procurement teams ask us first.
Does MuseMe train AI models on our collection?
No. Your collection is used only to prepare your stories. It is never used to train AI models, ours or anyone else's.
Can we delete our content?
Yes. Delete any source or story at any time. If you leave MuseMe, we delete all your content at the end of your contract and confirm it in writing.
Who owns the stories?
You do. Your sources stay yours. The stories prepared from them belong to your institution, to use on any channel.
Where is visitor data stored?
In Europe. All MuseMe data is stored on European servers.
Read the full guide to AI in museums
Trust by Design explains the EU AI rules for cultural institutions, step by step.
Get the white paper