---
What you need
ClauseGuard is web-first. You don't install anything or wire up infrastructure to get your first report. You need two inputs: a plain-language description of your AI system (what it does, what data it touches, who it serves) and your target market (EU or Global). No API key is required for the demo path — the app runs in mock mode and returns a representative output you can review immediately.
Three steps
- Describe the system — Paste a short description: "An LLM chatbot that summarizes user emails and suggests replies, trained on user data." Be honest about data and human involvement; they change the tier.
- Pick the market — Choose EU or Global. Choosing EU turns on the full EU AI Act obligation set (Annex III, Art. 9–17, Art. 50).
- Score it — ClauseGuard returns a risk-tier classification and a gap report listing, per article, what you can and cannot yet evidence.
Reading your readiness report
The report is a view of exposure. The risk tier tells you which obligation set attaches. The gap list tells you, per article, what to close and who should own it. Treat each "gap" as a remediation task, not a verdict. Export the report and attach it to your conformity file as evidence that you classified and assessed the system.
What happens next
After the first run, teams typically assign owners to each gap, re-score after changes, and keep the report in the conformity file. ClauseGuard is the fastest path from "we ship AI into the EU" to "we know our gaps" — it is not a conformity certificate and does not replace counsel. When the system changes materially, score again.
Authoritative references
- EU AI Act (Reg. 2024/1689), Art. 6 / 9 / 14 / 50: https://eur-lex.europa.eu/eli/reg/2024/1689/oj
- AI Act explorer: https://artificialintelligenceact.eu/