In one sentence: complying with it, and being able to prove it to your clients.
Hello, has my order been shipped?
Yes, it was shipped this morning. Would you like the tracking link?
For an SME that uses AI tools, the regulation mainly asks for three things: transparency towards people, competence within teams, and caution on a few sensitive uses.
The people you deal with, when they are interacting with an AI — on your website, on the phone, by email.
Content generated by an AI.
In your teams, on the tools they use.
To screen job applications or assess your employees.
So as not to stray into them unknowingly.
For most SMEs, the question will not come from an inspection, but from a main contractor asking you, in a call for tenders or a questionnaire, for guarantees on your uses of AI. This is already happening with other security regulations. Penalties exist; for an SME, the lower of the two ceilings provided for applies. We do not make it a selling point: what matters is being able to answer your clients.
The regulation applies in stages. This page describes the law as at 11 September 2026; we review it every six months.
Sources: Regulation (EU) 2024/1689 on artificial intelligence (official text); Regulation (EU) 2026/1744, which postpones some of its deadlines.
Who decides, who approves, and how you stay in control of your agents.
Where your data goes, and how to require that it stays in France.
Personal data handled by the rules, from the design stage.
The schemes that can reduce what you pay, and the application we prepare with you.
Our free tool to choose an AI model by task, price and sovereignty.
The AI Usage Audit takes stock of your uses, identifies what needs fixing and gives you a file of evidence. Not sure? The free Flash Diagnostic points you in the right direction.