PricingLog in

AI Act timeline and key dates.

Reviewed

Short answer

The AI Act entered into force on 1 August 2024 and applies in stages: prohibitions and AI literacy from 2 February 2025, general-purpose AI model rules from 2 August 2025, most other rules, including Article 50, from 2 August 2026, and high-risk rules from 2 December 2027 (Annex III) and 2 August 2028 (Annex I), as set by Regulation (EU) 2026/1744. Generative AI systems placed on the market before 2 August 2026 have until 2 December 2026 to mark their outputs, the same day two new prohibitions start to apply.

Official sources: AI Act, Article 113 (consolidated) · AI Act, Article 111 (consolidated)

to the end of 2 December 2026, Central European Time (our reading of “by 2 December 2026”).

Every date

AI Act dates, consolidated text as at 27 July 2026
DateWhat appliesProvision
12 July 2024The AI Act is published in the Official JournalRegulation (EU) 2024/1689
1 August 2024Entry into force (twentieth day after publication)Article 113
2 February 2025General provisions, including AI literacy, and the prohibited practicesArticle 113(a)
2 August 2025Notified bodies, general-purpose AI models, governance, penalties (except fines on GPAI providers) and confidentialityArticle 113(b)
24 July 2026The Digital Omnibus on AI is published in the Official JournalRegulation (EU) 2026/1744
27 July 2026The Omnibus enters into force: new wording of Article 4, new SMC reliefs; Articles 102–110 applyRegulation (EU) 2026/1744, Article 4; Article 113(d)
2 August 2026General date of application: Article 50 transparency rules, enforcement, Commission fines on GPAI providersArticle 113; Article 101
2 December 2026Marking duty for generative AI systems placed on the market before 2 August 2026; bans on AI-generated non-consensual intimate imagery and child sexual abuse materialArticle 111(4); Article 5(1)(ba), (bb)
2 August 2027General-purpose AI models placed on the market before 2 August 2025 must comply; each member state has at least one AI regulatory sandbox runningArticle 111(3); Article 57(1)
2 December 2027High-risk rules for Annex III systems (employment, education, credit, insurance, law enforcement and more)Article 113(c)(i)
2 August 2028High-risk rules for AI in products covered by Annex I (for example toys, lifts or medical devices)Article 113(c)(ii)
2 August 2030High-risk systems intended for public authorities, already on the market, must complyArticle 111(2)
31 December 2030AI components of large EU IT systems (Annex X) placed on the market before 2 August 2027 must complyArticle 111(1)

High-risk systems already on the market before their application date are covered only if their design changes significantly from that date, except systems for public authorities, which must comply by 2 August 2030 (Article 111(2)).

The next deadline: 2 December 2026

  • Providers of AI systems, including general-purpose AI systems, that generate synthetic audio, image, video or text and were placed on the market before 2 August 2026 must comply with Article 50(2) by this date: outputs marked in a machine-readable format and detectable as AI-generated (Article 111(4)).
  • The new prohibitions in Article 5(1)(ba) and (bb) apply from the same day.
  • Deployers’ duties under Article 50(4), such as disclosing deepfakes, already apply since 2 August 2026; there is no transition for them, and content generated before 2 August 2026 does not need retroactive labels.
  • Who the deadline covers and what a provider does before it, step by step: 2 December 2026: the AI Act marking deadline.

What AI tools document today

Article 50(2) is about providers marking their outputs. This table shows what each tool’s own documentation says about C2PA Content Credentials and watermarks. It is a record of documents, not a finding about anyone’s compliance; each row links to the page with the vendor’s sentence and source.

AI tools and what their documentation says about Content Credentials and watermarks
ChatGPT, Codex and the OpenAI API (images)Documented: addsDocumented: adds
Sora (video; discontinued)Documented: addsNot documented
Adobe FireflyDocumented: addsNot documented
Google Gemini image models, Imagen and VeoDocumented: addsDocumented: adds
Microsoft DesignerDocumented: addsNot documented
Bing Image CreatorDocumented: addsNot documented
Meta AI image generationNot documentedDocumented: adds
Stability AI (hosted API)Documented: addsNot documented
FLUX (Black Forest Labs)Documented: addsDocumented: adds
MidjourneyNot documentedNot documented
Amazon Nova CanvasDocumented: addsDocumented: adds
RunwayDocumented: addsDocumented: adds
Grok Imagine (xAI)Not documentedDocumented: visible watermark
ElevenLabsNot documentedDocumented: adds
SynthesiaDocumented: addsNot documented

Questions

When does the Article 50(2) transition end?
On 2 December 2026. Article 111(4) of the AI Act, as added by Regulation (EU) 2026/1744, says providers of AI systems generating synthetic audio, image, video or text content that were placed on the market before 2 August 2026 shall take the necessary steps to comply with Article 50(2) by 2 December 2026.
Were the high-risk rules delayed?
Yes. Regulation (EU) 2026/1744 moved them from 2 August 2026 to 2 December 2027 for Annex III systems, and from 2 August 2027 to 2 August 2028 for AI in products covered by Annex I.
What is the penalty for not complying with Article 50?
Article 99(4)(g) lists the Article 50 transparency obligations among infringements subject to fines of up to EUR 15 000 000 or, for an undertaking, up to 3 % of total worldwide annual turnover, whichever is higher. For SMEs, including start-ups, and for small mid-cap companies, the lower of the two amounts is the cap.

Sources

Last reviewed by Volkan Çatak, Founder, Proofcan.

Not legal advice. This page explains official texts in plain words and links each one; check the sources, or a lawyer, before relying on it for a decision.