Skip to content

Comply with the AI Act with full autonomy, one phase at a time

Risk classification, system registry, use policy, human oversight and technical documentation. Everything mapped article by article to EU Regulation 2024/1689.

⏱ 8–12 weeks 6 guided phases 4 key resources

Does this sound familiar?

🤖

You use AI (ChatGPT, copilots, in-house models) without knowing which AI Act risk category you fall into.

📋

There's no registry of which AI systems each team uses, or who approved them.

🚦

Nobody has defined what's allowed and what isn't when using AI with customer data.

📁

If a regulator asked tomorrow, you wouldn't have a single document to show.

What AI Act risk level are you at?

Answer 4 questions and get an instant, indicative classification.

Quick risk classifier
Indicative only. Not a substitute for legal advice.

1. Does your AI system subliminally manipulate people's behaviour, or is it used for social scoring?

2. Does your system make or decisively influence decisions about people: hiring, credit, health, justice?

3. Is it a chatbot, content generator, or does it interact directly with people simulating a human?

4. Do you use it as internal support (analytics, productivity, internal task automation)?

Indicative result generated by simple rules, without recording your answers. For a formal classification, see the full assessment or consult a professional.

One framework, four risk levels

It's the framework of roles, policies and controls that ensures AI use in your company is safe, transparent and AI Act-compliant — from the simplest chatbot to a high-risk system.

Unacceptable risk · Prohibited
High risk
Limited risk · Transparency
Minimal risk
2024
Regulation enters into force
2025
Prohibitions and AI literacy
2026
Transparency obligations
2027
High-risk systems (Annex III)

Reference timeline following the May 2026 Digital Omnibus. Always check the latest official update.

6 guided phases, with a deliverable in each

Click each phase to see what you'll do and which template you'll get.

Choose your profile

The same result, with full control in your hands

Dimension Traditional approach AI Governance path
Cost €10,000 – €40,000+ From a single template to the full pack
Time Months of back-and-forth 8–12 weeks, at your own pace
Approach Generic legal report Ready-to-use operational templates
Updates Pay again for every regulatory change Content reviewed as the timeline progresses
120+
in estimated external legal-audit cost avoided (approx. €1,200/system)

Ready-to-fill templates — buy now

📋
Inventory · Art. 6 + Annex III
AI Act Checklist — What your company needs
32 controls across 4 blocks: inventory, risk levels and minimum documentation. Excel + PDF.
Classification · Annex III
AI Act Risk Classification Checklist
Interactive Excel decision tree to classify each system with documented justification.
📄
Policy · Art. 5 + Art. 50
Acceptable AI Use Policy
Editable Word document: Art. 5 prohibitions, Art. 50 transparency, and generative-AI use.
Deadlines · Annex III · Dec 2027
AI Act Regulatory Deadlines Checklist
A 90-day plan with 32 prioritised controls and an automatic progress dashboard.
FULL PACK · 9 PRODUCTS Save €102

The whole catalogue in one download

AI Act · Data Governance · NIS2 · GDPR · ISO 42001. Also includes the resources from the Data Governance path.

€299 €401 · VAT included · one-time payment
Buy the pack → Instant download
No subscription

See all resources for this path in the store →

Related articles

📅
Timeline
What exactly changes and when, article by article.
👤
Roles
What they do, and why it fits Phase 2 of this path.
🏛️
Committee
The decision-making structure that underpins this whole path.
⚖️
Frameworks
Complements Phase 5 (technical documentation) of this path.

See all articles →

Before you ask

Does the AI Act apply to me if I only use ChatGPT or Copilot internally?
Probably yes, although with limited obligations (mainly transparency and AI literacy). The exact level depends on how you use it: try the risk classifier on this page.
What happens if I don't comply?
The Regulation provides for penalties that can reach several million euros, or a percentage of global turnover, depending on the severity of the infringement.
Does this replace legal advice?
No. It's the operational and documentary foundation; for specific legal interpretations of your case, consult a professional. Full legal notice →
Do I need to sort out my Data Governance first?
It's not mandatory, but if you're training or fine-tuning models with your own data, Art. 10 requires governance of that data. We also recommend reviewing the Data Governance path.

Discover your AI Act exposure in 12 questions

Free, no sign-up. We'll tell you which obligations apply now and which come later.