AI Transparency
The AI phrases and explains. It does not invent prices, availability, opening hours or entry requirements — those always come from a verified source.
Last updated:
Draft pending legal review — this text is not final and has not yet been approved by a lawyer.
What the AI does
A language model is used where understanding or phrasing is needed, never to establish facts.
- Understands your free-text request and turns it into structured fields.
- Asks one clarifying question when a material detail is missing.
- Explains why a plan was chosen and what was traded away.
- Writes the day narrative and the budget summary.
What the AI does not do
These rules are enforced in code, not in a prompt instruction.
- It never sets a price, availability or cancellation terms — those come from the supplier with a timestamp.
- It never asserts opening hours; where there is no verified source, the copy says the hours must be checked.
- It never determines visa or health requirements — we show the source, the date and a pointer to official verification.
- It never receives payment details, identity documents or system secrets.
How we verify
Every plan passes a rules engine and validation before you see it: budget fit, time feasibility, distances and opening hours.
Every claim shown to you is linked to a source record with the time it was checked. Output that fails the schema or the rules is rejected and never displayed.
Your data and the model
We do not train models on your data without explicit consent and a dedicated policy.