The EU AI Act (Regulation (EU) 2024/1689) regulates AI systems placed on the EU market according to risk. It applies in phases: prohibited practices since February 2, 2025; general-purpose AI model obligations since August 2, 2025; and Article 50 transparency obligations from August 2, 2026. The Digital Omnibus on AI (Regulation (EU) 2026/1744, in force since July 27, 2026) deferred the high-risk system obligations: stand-alone Annex III systems now apply from December 2, 2027 and AI embedded in Annex I regulated products from August 2, 2028.
What does EU AI Act actually require?
Key obligations include: Classify each AI system against the Act's risk tiers (prohibited, high-risk per Annex III, limited-risk, or minimal-risk) and document the determination.; For high-risk systems: implement a risk management system, data governance, technical documentation, automatic logging, transparency to deployers, human oversight, and accuracy/robustness/cybersecurity measures.; Register stand-alone high-risk systems in the EU public database before placing them on the market.; Conduct a conformity assessment (self-assessment for most Annex III systems; notified-body assessment for biometric and certain product-embedded systems) and obtain CE marking.; Providers of general-purpose AI models: publish a training-data summary, comply with EU copyright law, and (for models with systemic risk) conduct adversarial testing and report serious incidents.; Deployers of high-risk systems: ensure human oversight, monitor operation, retain logs for at least six months, and conduct a fundamental-rights impact assessment for public bodies and certain private actors..
Who is in scope of EU AI Act?
EU AI Act (EU) — status: Phased — in force; high-risk rules deferred. Scope attaches based on jurisdiction and the role a company plays in the AI supply chain. See /frameworks/eu-ai-act for the full scope note and source links.
When does EU AI Act take effect?
Phased: prohibitions from Feb 2, 2025; GPAI from Aug 2, 2025; Art. 50 transparency from Aug 2, 2026; high-risk (Annex III) from Dec 2, 2027; high-risk in regulated products (Annex I) from Aug 2, 2028. Status reviewed 2026-09-27; see the framework brief for all key dates and sources.
What are the penalties?
Maximum penalties: Tiered (Art. 99): up to €35M or 7% of global turnover for prohibited practices (Art. 5); up to €15M or 3% for high-risk and GPAI obligation violations; up to €7.5M or 1% for supplying misleading information. SMEs/start-ups: lower of the two thresholds (Art. 99(6)).. Enforcement is carried out by the designated authorities in the jurisdiction.
Which vendors help with EU AI Act compliance?
In our directory, the following vendors reference EU AI Act in their compliance coverage: Credo AI, Holistic AI, Fiddler AI, Robust Intelligence, CalypsoAI, Monitaur, Trustible, FairNow, Fairly AI, Saidot, LatticeFlow AI, Lakera. Each profile links to the public source for the claim.