How the assessment works
A structured, deterministic interpretation of the EU AI Act risk framework translated into an informational self-assessment model.
No generative AI is used to determine your risk classification.
From regulation to decision logic
The EU AI Act introduces a risk-based classification system for AI systems, but its application can be complex in practice.
This tool translates the EU AI Act framework into a structured set of decision rules, allowing a fast and consistent informational assessment.
Methodology in 3 steps
Framework mapping
We map the EU AI Act provisions into a structured set of classification criteria.
- —Articles and definitions are analyzed and grouped into decision-relevant elements
- —Key conditions (use case, context, impact) are identified
- —Ambiguities are handled with conservative interpretation logic
Decision model
The criteria are translated into a deterministic decision framework.
- —Questions are designed to capture legally relevant factors
- —Each answer activates specific logical branches
- —The system follows a predefined classification path
Risk indication
The output provides a preliminary risk classification with contextual explanation.
- —Likely risk category (e.g. prohibited, high-risk, limited risk)
- —Key drivers behind the classification
- —Directional indication for next steps
What this assessment is designed for
- ✓Early-stage informational screening
- ✓Internal decision support
- ✓General understanding before professional review
What this tool is and is not
What it is
- ✓A structured first-pass classification based on the EU AI Act
- ✓A consistent and repeatable classification approach
- ✓A support tool for internal product review
What it is not
- ×Legal advice
- ×A formal approval or official decision
- ×A substitute for a full compliance assessment
Design principles
Clarity over complexity
The framework logic is simplified without losing essential meaning.
Determinism over opacity
The same inputs always produce the same output.
Conservative interpretation
When ambiguity exists, the model prioritizes safer classification paths.
Limitations
The assessment is based solely on the information provided by the user and does not account for:
- —Full system architecture or technical implementation details
- —Jurisdiction-specific interpretations or future regulatory updates
- —Contextual factors not explicitly captured in the questionnaire
Part of a broader informational toolkit
This tool is the first component of a broader set of EU AI Act informational tools designed to support organizations across different stages:
- —Initial classification
- —Requirements overview and documentation topics
- —Ongoing informational review
Start your assessment
Get a structured indicative classification in minutes and understand where your AI system may fall under the EU AI Act.
Start assessmentFree assessment - 2-3 minutes - No account required