This AI Policy explains how Rezvion (“Rezvion,” “we”) uses artificial intelligence in the Service, what you can and cannot rely on AI output for, and how we protect your data when AI features are involved. It supplements our Privacy Policy and Terms of Service.
1. AI can make mistakes — always review before acting
Rezvion’s AI features (including the in-app AI support assistant and AI-generated content) can produce output that is inaccurate, incomplete, or out of date. AI responses are suggestions, not instructions. You must review any AI output and verify important details before you act on it — especially anything affecting payments, payouts, refunds, taxes, contracts, food safety, allergens, or legal and financial matters. When in doubt, contact a human at partnerships@rezvion.nl.
2. Where we use AI
- AI support assistant — answers questions about the platform using our help-center articles, and hands off to a human when it is unsure or the topic involves money.
- Content generation — drafts such as menu descriptions, marketing copy, social captions, ad copy, images, and video, which you review and edit before publishing. AI-generated images and video can contain visual errors or artefacts — preview them and confirm they represent your business accurately before running them as ads.
- Operational assistance — features such as supplier invoice extraction and demand forecasting, presented for your review.
AI output is generated to assist you. You remain responsible for what you publish, send, or act upon.
AI image and video generation is metered: each plan includes a monthly allowance, with pay-per-use pricing once you go over it. See Pricing for current allowances.
3. Human oversight
For sensitive topics — money, payouts, refunds, billing, and contract terms — the AI assistant is designed to lower its confidence and route the conversation to a human on the Rezvion team rather than answer on its own. You can also ask to speak to a human at any time.
4. AI-generated content & transparency (EU AI Act)
In line with the EU AI Act's transparency rules:
- You always know when you're talking to AI. The support assistant and in-app AI helpers clearly identify themselves as AI.
- AI-generated images and videos are disclosed as artificially generated when a restaurant publishes them through Rezvion to social media or ads.
- We run no “high-risk” or prohibited AI. Our AI assists with content, analytics, and support — it does not make automated decisions about people (no biometric, social-scoring, or individual worker-evaluation use).
5. How we protect your data
We apply guardrails before any prompt is sent to an external AI provider:
- Personal data is stripped. Before a prompt reaches an AI provider, we automatically redact personal and payment details we detect — such as email addresses, phone numbers, IBANs, card numbers, and dates of birth. Please do not share payment card numbers, ID numbers, or other sensitive personal data with the AI assistant.
- Content safety + scope. AI input and output are checked for unsafe content, and the assistant only answers questions about Rezvion and restaurant operations.
- No sensitive data is stored in our AI audit log. We keep a log of guardrail events (e.g. that a redaction occurred) for safety and quality, but it records only event types and counts — never the content of your prompts or any personal data.
We use third-party AI providers (for example OpenAI) as processors to deliver these features. Restaurants remain the data controller for their own customers’ data; see our Sub-processors list and Data Processing Agreement for details.
6. Not professional advice
AI output is not legal, financial, tax, medical, or food-safety advice. It must not be relied upon as a substitute for professional judgement or for verifying your own records and obligations.
7. Your responsibilities
- Review and verify AI output before acting on or publishing it.
- Do not enter sensitive personal or payment data into AI features.
- Do not attempt to misuse AI features (for example, to generate unlawful, deceptive, or harmful content) — see our Acceptable Use Policy.
8. Changes to this policy
We may update this AI Policy as our AI features evolve. The “Last updated” date above reflects the most recent change.
9. Contact
Questions about how we use AI? Email partnerships@rezvion.nl.