EU AI Act
Can you pass this 10-question EU AI Act quiz?
Ten multiple-choice questions on the EU AI Act as amended in July 2026: the dates, the fines and the duties people get wrong. Tap an answer to see why, with the Article. Takes about five minutes.
Can you get 10 out of 10?
1 of 10Since when have the Article 5 bans applied?
Show answer
Answer: B. Chapters I and II, which include Article 5, have applied since 2 February 2025. Source: EU AI Act, Article 113 as amended by Regulation (EU) 2026/1744
2 of 10From when do the duties for Annex III high-risk systems apply?
Show answer
Answer: C. The July 2026 amendment moved them from 2 August 2026 to 2 December 2027. Source: EU AI Act, Article 113 as amended by Regulation (EU) 2026/1744
3 of 10From when do the high-risk duties apply to Annex I products?
Show answer
Answer: D. Annex I systems follow on 2 August 2028. Source: EU AI Act, Article 113 as amended by Regulation (EU) 2026/1744
4 of 10Since when have the rules for general-purpose AI models applied?
Show answer
Answer: B. Chapter V has applied since 2 August 2025. Source: EU AI Act, Article 113 as amended by Regulation (EU) 2026/1744
5 of 10What is the top fine for breaching an Article 5 ban?
Show answer
Answer: C. Up to EUR 35 million or 7% of worldwide annual turnover, whichever is higher. Source: EU AI Act, Article 99(3)
6 of 10Which fine tier covers a provider's Article 16 obligations?
Show answer
Answer: B. Article 99(4)(a) puts the Article 16 obligations in the middle tier. Source: EU AI Act, Article 99(4)
7 of 10How long must a deployer of a high-risk system keep its automatic logs, at least?
Show answer
Answer: C. At least six months, unless other law says otherwise. Source: EU AI Act, Article 26(6)
8 of 10Which of these is banned?
Show answer
Answer: B. Emotion inference in the workplace and in education is banned, except for medical or safety reasons. Source: EU AI Act, Article 5(1)(f)
9 of 10Which Annex III area is excluded from the fundamental rights impact assessment?
Show answer
Answer: C. Article 27(1) excludes systems in Annex III, point 2, critical infrastructure. Source: EU AI Act, Article 27(1)
10 of 10Who must disclose a deepfake under Article 50(4)?
Show answer
Answer: B. Article 50(4) puts the disclosure duty on the deployer, from 2 August 2026. Source: EU AI Act, Article 50(4)
EU AI Act flashcards for Anki
front,back Since when have the Article 5 bans applied? | A) 1 August 2024 / B) 2 February 2025 / C) 2 August 2026 / D) 2 December 2027,"B) 2 February 2025. Chapters I and II, which include Article 5, have applied since 2 February 2025." From when do the duties for Annex III high-risk systems apply? | A) 2 August 2026 / B) 2 February 2027 / C) 2 December 2027 / D) 2 August 2028,C) 2 December 2027. The July 2026 amendment moved them from 2 August 2026 to 2 December 2027. From when do the high-risk duties apply to Annex I products? | A) 2 August 2026 / B) 2 December 2027 / C) 2 August 2027 / D) 2 August 2028,D) 2 August 2028. Annex I systems follow on 2 August 2028. Since when have the rules for general-purpose AI models applied? | A) 2 February 2025 / B) 2 August 2025 / C) 2 August 2026 / D) 2 December 2027,B) 2 August 2025. Chapter V has applied since 2 August 2025. What is the top fine for breaching an Article 5 ban? | A) EUR 7.5 million or 1% / B) EUR 15 million or 3% / C) EUR 35 million or 7% / D) EUR 50 million or 10%,"C) EUR 35 million or 7%. Up to EUR 35 million or 7% of worldwide annual turnover, whichever is higher." Which fine tier covers a provider's Article 16 obligations? | A) EUR 35 million or 7% / B) EUR 15 million or 3% / C) EUR 7.5 million or 1% / D) No fine,B) EUR 15 million or 3%. Article 99(4)(a) puts the Article 16 obligations in the middle tier. "How long must a deployer of a high-risk system keep its automatic logs, at least? | A) One month / B) Three months / C) Six months / D) Two years","C) Six months. At least six months, unless other law says otherwise." "Which of these is banned? | A) A chatbot that tells users it is AI / B) Inferring employees' emotions at work, for productivity / C) A spam filter / D) Credit scoring with human review","B) Inferring employees' emotions at work, for productivity. Emotion inference in the workplace and in education is banned, except for medical or safety reasons." Which Annex III area is excluded from the fundamental rights impact assessment? | A) Employment / B) Education / C) Critical infrastructure / D) Essential services,"C) Critical infrastructure. Article 27(1) excludes systems in Annex III, point 2, critical infrastructure." Who must disclose a deepfake under Article 50(4)? | A) The provider of the model / B) The deployer who publishes it / C) The platform that hosts it / D) Nobody,"B) The deployer who publishes it. Article 50(4) puts the disclosure duty on the deployer, from 2 August 2026."
Where next?
Also free: the EU AI Act interview questions and the EU AI Act mind map, plus every other credential on our study tools page. For depth, the AI Governance Framework and the AI Governance Worked Scenarios trace each duty to its Article, written to the 2024 text.
Frequently asked questions
Is this EU AI Act quiz like the real exam?
No. It is a 10-question recall check written by Credential Press, not exam questions. Each answer cites its source.
Can I study the EU AI Act quiz offline?
Yes. Download the free flashcard CSV and import it into Anki or any flashcard app that reads CSV.
Since when have the Article 5 bans applied?
2 February 2025. Chapters I and II, which include Article 5, have applied since 2 February 2025.
Which books go deeper on EU AI Act?

Building one that survives the EU AI Act. 22 chapters, 384 pages.

AI Governance Worked Scenarios
24 worked situations under the EU AI Act, decided against the Articles. 81 pages.
Sources
- Regulation (EU) 2024/1689 (AI Act), consolidated text of 27 July 2026, read 2 October 2026
- Regulation (EU) 2026/1744 (Digital Omnibus on AI), read 2 October 2026
Credential Press is independent of the EU institutions.