Regulation (EU) 2024/1689 · Digital Omnibus 2026
Fines up to €35M are enforceable. Find out in 10 minutes if you're exposed.
"We're not in Europe" is the most expensive assumption in AI right now — if your outputs reach EU users, the Act reaches you. Classify every AI system, get your exact risk tier, binding deadline, and penalty exposure — plus an action plan written for your systems, not a generic checklist.
Transparency obligations (chatbots, deepfakes)
On scheduleAnnex III high-risk obligations
Deferred +16moAnnex I high-risk (medical, machinery, vehicles)
Deferred +12moPer system
risk tier, deadline and penalty band
~15 min
self-service, no sales call
PDF + link
shareable report with an action plan
ISO 42001
aligned methodology
The Problem
Compliance starts with one question: which systems, which tier?
Get it wrong and you either over-engineer a minimal-risk tool — or miss binding obligations on a high-risk one.
Are we even in scope?
The Act binds providers and deployers alike — and reaches non-EU companies whose AI outputs are used in the EU. Most organizations have never mapped their territorial exposure.
Which category are we in?
Classification is per-system, not per-company. A chatbot, a credit model, and an HR screening tool can each land in a different tier — with entirely different obligations.
The deadlines keep moving.
The 2026 Digital Omnibus deferred high-risk obligations to Dec 2027 / Aug 2028 — yet transparency rules still take effect August 2026. Generic guidance is already out of date.
The Deliverable
A complete, system-by-system readiness report
Produced in about ten minutes — no spreadsheets, no legal retainer.
Per-system risk classification
Each AI system mapped to Prohibited, High-Risk (Annex III/I), Limited, or Minimal risk.
Tailored obligations checklist
The exact Articles and duties per system — risk management, documentation, human oversight, transparency.
Your deadline countdown
The binding compliance date for every system, using the updated 2026 Digital Omnibus timeline.
Penalty exposure estimate
The maximum fine band you face per system — up to €35M / 7% of global turnover.
Governance readiness score
A 0–100 benchmark across inventory, oversight, documentation, and vendor controls.
Prioritized action plan
Where to begin first, based on highest-risk systems and earliest binding deadlines.
The Timeline
The updated EU AI Act calendar
Reflecting the Digital Omnibus on AI (May 2026) — high-risk obligations were deferred, but transparency rules still apply from August 2026.
AI Act enters into force
Article 113. No change.
Prohibited practices + AI literacy
Article 5 bans & Article 4 literacy obligations apply.
GPAI model obligations
Articles 51–56. AI Office operational.
Article 50 transparency obligations
Chatbots, deepfakes, synthetic content disclosure. Still on schedule.
Watermarking grace ends + new NCII/CSAM ban
Grace period for existing systems ends; new Article 5 prohibition applies.
National regulatory sandboxes deadline
Deferred 1 year by the Omnibus.
Annex III high-risk obligations apply
DEFERRED from 2 Aug 2026 (+16 months).
Annex I high-risk obligations apply
DEFERRED from 2 Aug 2027 (+1 year). Medical devices, machinery, vehicles.
Deadlines verified as of July 2026 · Sources: Regulation (EU) 2024/1689 (EUR-Lex) and the May 2026 Digital Omnibus on AI.
Begin
Know exactly where you stand on the EU AI Act
Classify your AI systems, get your obligations and deadlines, and see your penalty exposure — at no cost.
This assessment provides guidance based on publicly available information about the EU AI Act and is not legal advice. Final dates take legal effect upon publication in the Official Journal.