AI Intensify: Practical AI for small business owners, explained plainly, with the real costs.

© 2026 AI Intensify

Europe · EU

European Union

Partly in force Legally binding Checked
All regions

In short

  1. Users must support staff AI literacy and avoid banned practices.
  2. Published deepfakes and some public-interest AI text need clear disclosure.
  3. High-risk uses like hiring carry extra duties from December 2027.
Not legal advice: These pages summarize official sources in plain English and are not legal advice. For decisions about your own business, check the official text or ask a qualified lawyer where you operate.

What the rule is

AI Act, Regulation (EU) 2024/1689, amended by Regulation (EU) 2026/1744 (Digital Omnibus on AI)

Where it stands

Partly in force Legally binding

Since : Entry into force of the AI Act (most rules apply from 2 August 2026)

Who it applies to

Providers (who develop an AI system or model, or have it developed, and place it on the EU market under their own name), deployers (anyone using an AI system under its authority in a professional activity), importers and distributors. It also covers non-EU companies whose AI output is used in the EU.

What it asks of businesses

  • Take measures to support AI literacy of staff and contractors who use AI. No certificate is needed; organizations can keep an internal record of trainings and other initiatives (Commission Q&A).
  • Do not use banned AI practices, such as manipulative techniques causing harm, social scoring, emotion recognition at work or in education (except for medical or safety reasons), or biometric categorization that infers traits such as race, political opinions or sexual orientation.
  • Disclose deepfake images, audio or video you publish (lighter disclosure for clearly artistic, satirical or fictional work).
  • Disclose AI-generated text published to inform the public on matters of public interest, unless it had human review or editorial control and someone holds editorial responsibility.
  • Inform people exposed to emotion recognition or biometric categorization systems.
  • Disclose clearly at the first interaction or exposure (Art. 50(5)).
  • If you use AI for high-risk purposes such as screening job applicants: deployer duties under Art. 26 (use per instructions, human oversight, monitoring, logs, informing workers) apply from 2 December 2027. Deployers using AI to assess creditworthiness or to price life and health insurance must also assess the impact on fundamental rights (Art. 27).
  • If you build or brand an AI system (provider): design AI that interacts with people so they are informed it is AI (Art. 50(1)) and add machine-readable marking to synthetic content (Art. 50(2)).

Key dates

  1. Since : Entry into force of the AI Act (most rules apply from 2 August 2026)

  2. Since : Definitions, AI literacy duty and banned AI practices (Chapters I and II)

  3. Since : Rules for general-purpose AI models (Chapter V)

  4. Since : Governance rules and penalties chapter

  5. Since : Entry into force of the Digital Omnibus on AI (Regulation (EU) 2026/1744) and the amended AI literacy rule (Art. 4)

  6. Since : General application date, including Art. 50 transparency duties for deployers (Art. 50(3) and 50(4))

  7. Since : Supervision and enforcement by national authorities

  8. Since : Provider duties: AI that interacts with people must tell them it is AI (Art. 50(1)), and generative AI systems placed on the market from this date must mark their output (Art. 50(2))

  9. From : New bans on AI that generates non-consensual intimate content or child sexual abuse material, and on nudification apps

  10. From : Generative AI systems placed on the market before 2 August 2026 must have machine-readable marking (Art. 50(2))

  11. From : General-purpose AI models placed on the market before 2 August 2025 must comply (Art. 111(3))

  12. From : High-risk rules for Annex III uses such as hiring, credit, education and essential services, including deployer duties

  13. From : High-risk rules for Annex I systems in regulated products such as machinery, toys and lifts

Penalties

Banned practices: up to EUR 35 million or 7% of worldwide annual turnover. Most other breaches, including Art. 50 transparency: up to EUR 15 million or 3%. Misleading information to authorities: up to EUR 7.5 million or 1%. Large firms pay whichever is higher; SMEs and start-ups pay whichever is lower (Art. 99(6)), extended to small mid-caps for Art. 99(4) and (5) fines. Art. 4 AI literacy has no fine tier in Art. 99; any sanction depends on national law.

Who enforces it

National market surveillance authorities in each Member State; European Commission AI Office (general-purpose AI models). The European AI Board advises and coordinates; it does not enforce.

Official sources

  1. Publications Office of the European Union (EUR-Lex)eur-lex.europa.eu (opens in a new tab)
  2. Publications Office of the European Union (EUR-Lex)eur-lex.europa.eu (opens in a new tab)
  3. European Commission (AI Act Service Desk)ai-act-service-desk.ec.europa.eu (opens in a new tab)
  4. European Commission (AI Act Service Desk)ai-act-service-desk.ec.europa.eu (opens in a new tab)
  5. European Commission (AI Act Service Desk)ai-act-service-desk.ec.europa.eu (opens in a new tab)
  6. European Commission (AI Act Service Desk)ai-act-service-desk.ec.europa.eu (opens in a new tab)
  7. European Commission (AI Act Service Desk)ai-act-service-desk.ec.europa.eu (opens in a new tab)
  8. European Commissiondigital-strategy.ec.europa.eu (opens in a new tab)
  9. European Commissiondigital-strategy.ec.europa.eu (opens in a new tab)
All regions

How we check this

Each entry is built mainly from official sources, such as the law itself, government pages and regulator guidance, and those sources are linked on each jurisdiction page and on the timeline. The date on each entry shows when it was last checked against them. Dates that could not be confirmed on an official source are labeled Not yet confirmed, other unconfirmed details are left out, and dates that may still move are marked Provisional. Laws change, so the official text always comes first.