Skip to content
Orya AI Act

Find out what the EU AI Act asks of your company.

List the AI tools you use and answer a few plain questions. You get each tool's risk class, what the law asks of you with the article behind it, and your documents. In Greek and English, in about fifteen minutes.

Free for your first AI systems. No card needed.

Try it. What do you use AI for?

High-risk (Annex III)

This use is listed as high-risk. Strict requirements apply from the date below. Applies from 2 December 2027.

What the law asks of you

  • Tell people that AI is involved in decisions about them

    People affected must be told a high-risk AI system is used, and can ask for an explanation.

    Art. 26(11), Art. 86

  • AI literacy for your team

    Take measures to support AI literacy among the staff who use or operate the system.

    Art. 4

Worked out live by the same rules you use in the app.

Five steps, about fifteen minutes

  1. List the AI you use

    Add every AI tool your company uses or builds, from ChatGPT to your website chatbot. Common tools come with suggested answers.

    You getAn AI register

  2. Classify each system

    Answer a few questions per system to see its risk class and what the law asks of you.

    You getA classification report for each system

  3. Tell people about AI

    Add the ready-made AI disclosures to your website, chatbot and content, then mark them done.

    You getReady-made disclosures, a public notice page and a website badge

  4. Train your people

    Send everyone who uses AI for you the short AI course, and keep the record.

    You getA training record and certificates

  5. Adopt an AI use policy

    Set your rules for using AI at work and share them with your staff in their AI course.

    You getAn AI use policy your staff acknowledge

The dates that matter

The AI Act applies in stages. These are the ones most companies need to know.

  1. In force

    2 February 2025

    Banned practices are off limits, and every company using AI must support its people's AI literacy.

    Art. 5, Art. 4

  2. In force

    2 August 2026

    Transparency duties apply: tell people when they talk to an AI, and label deepfakes.

    Art. 50

  3. Coming

    2 December 2026

    Two more bans apply, and generative AI already on the market must mark its output.

    Art. 5(1)(ba), (bb); Art. 50(2)

  4. Coming

    2 December 2027

    High-risk rules apply to the uses listed in Annex III, such as hiring and credit scoring.

    Art. 6(2), Annex III

Documents you can hand to anyone who asks

Each one downloads as PDF or Word, in Greek or English, and cites the article behind every statement.

  • AI system inventory registerPDF, Word
  • AI system classification reportPDF, Word
  • AI transparency kitPDF, Word
  • AI literacy recordPDF, Word
  • AI use policyPDF, Word

Know where you stand today.

Start with one AI tool. It takes a few minutes.

Start free