What is EU AI Act?
The EU AI Act entered into force on August 1, 2024 and applies in phases. It sorts AI systems into four risk tiers: unacceptable (banned), high (for example recruitment or credit scoring), limited (transparency duties for chatbots and synthetic media) and minimal. General-purpose AI models carry their own documentation and copyright duties. It reaches providers and deployers whose AI is used in the EU, wherever they are based. Penalties reach up to 35 million euros or 7% of global annual turnover. The timeline has shifted: transparency rules apply from August 2, 2026, while Annex III high-risk obligations are currently scheduled for December 2, 2027.
Who needs it
- Startups building AI products for EU customers
- Companies using AI in hiring, credit, education or other high-risk areas
What usually triggers it
- You launch an AI product in the EU
- A customer asks how your AI system is classified under the Act
How the process works
- Inventory your AI systems and classify them by risk
- Map the obligations that apply to your role
- Put governance, documentation and oversight in place
- Track the phased application dates
How Holam helps
We work out whether EU AI Act is what your customers actually need, and in what order alongside other frameworks. We then introduce you to a compliance platform and specialists that fit your stage, and stay involved until the result is in your hands. Holam receives a referral fee from providers we introduce when you engage them. You pay the provider directly, we add nothing to their price, and you remain free to choose another provider.
Frequently asked questions
Does it apply to non-EU companies?
Yes, if their AI systems are placed on the EU market or their outputs are used in the EU.