What happens if I miss the 2 August deadline?

Operating a non-compliant high-risk AI system after 2 August 2026 can result in fines up to 15 million euros or 3% of global turnover. However, regulators are expected to take a proportionate approach, particularly for SMEs making good-faith compliance efforts.

Do I need a conformity assessment?

Only if you develop or deploy a high-risk AI system listed in Annex III. If you only use general-purpose AI tools (ChatGPT, Claude, Copilot, etc.) as a business user, conformity assessment is the provider's responsibility, not yours.

Quick answer

High-risk AI obligations, quality management systems, conformity assessments, and technical documentation requirements all take effect on 2 August 2026. Use this checklist to prepare.

Updated June 2026 · MmowW AI Compliance

EU AI Act August 2026 Compliance Checklist: Everything Due by 2 August

What Takes Effect on 2 August 2026

The EU AI Act's core provisions for high-risk AI systems take effect on 2 August 2026. This includes: mandatory risk management systems (Article 9), data governance requirements (Article 10), technical documentation (Article 11), record-keeping (Article 12), transparency obligations (Article 13), human oversight mechanisms (Article 14), and accuracy and robustness standards (Article 15).

Additionally, Annex III high-risk categories become enforceable — meaning AI systems used in employment, creditworthiness assessment, education, law enforcement, and critical infrastructure must comply with all Chapter 2 requirements.

Your 45-Day Action Plan

Week 1-2: Complete your AI system inventory. Classify each system against Annex III categories. Identify any high-risk systems that require conformity assessment.

Week 3-4: For each high-risk system, begin preparing technical documentation per Article 11. Establish or verify your quality management system per Article 17. Contact notified bodies if third-party conformity assessment is required.

Week 5-6: Implement monitoring and record-keeping systems per Articles 12 and 61. Train relevant staff on human oversight requirements per Article 14. Finalise your risk management system per Article 9.

Common Mistakes to Avoid

Do not assume general-purpose AI tools like ChatGPT are high-risk — most are not. Do not confuse provider obligations with deployer obligations — as a business user, your obligations are primarily under Articles 26-27. Do not wait for national implementation — the regulation is directly applicable.

Do not over-invest in compliance for low-risk systems. Focus resources on any genuinely high-risk AI applications in your organisation. For most SMEs, the primary obligation remains Article 4 (AI literacy), which is already in force.

Check your AI compliance readiness — free.

Take the Readiness Check 3 minutes · 10 questions · no signup required

This article is for informational purposes only and does not constitute legal advice. Regulatory requirements change frequently — verify current rules with official sources. Built by Sawai Gyoseishoshi Office, Hiroshima, Japan.