Compliance Assistant - GDPR and EU AI Act
About this project
An AI-powered assistant that answers “can we build this?” for product teams by turning a plain-language description of a planned feature into an article-cited GDPR / EU AI Act assessment.
What it does
Describe a feature in plain words: for example, “we want to add an AI chatbot to our website for customer service” or “we want AI to screen job applications”.
Get a verdict from the law: the assistant reads the GDPR and the EU AI Act, and answers with proceed, proceed with conditions, or do not build as described.
Findings, each backed by the law: a finding is something to fix or fulfil, such as “tell users they are talking to an AI” or “a human must review every rejection”. Each finding names the exact article it rests on, and every citation is checked against the law’s text before you see it.
Human oversight and learning from feedback: nothing becomes an official record until the organisation’s Data Protection Officer approves it. The DPO can also send the feature back for changes, and the assistant learns from every decision.
Try it → compliance-web-qo1l.onrender.com
Sign in with your Google account. You will be a product manager who wants to check whether a newly planned feature is compliant with the GDPR and the EU AI Act. Describe a feature and read the assessment. The demo runs on a free tier and may take a minute to wake up; please enter no real personal data.
How the app itself stays compliant
Personal data never reaches a model: e-mail addresses, phone numbers, and card numbers are removed before any AI model sees the text.
Zero data retention: the AI providers keep nothing and do not train on requests.
Data stays in the EU, and not for long: the database is in Frankfurt, and finished conversations are deleted after 30 days.
Roles and a public privacy notice: each role sees only its own screens and tools, and the privacy notice explains what goes where.