Are all AI tools in media high-risk?

No. Only specific applications listed in Annex III are high-risk. General-purpose AI tools used for routine tasks (email, scheduling, document drafting) are typically not high-risk, even in regulated sectors like media.

When must we be compliant?

Article 4 (AI literacy) is already in force. High-risk AI system requirements take effect 2 August 2026.

Quick answer

The EU AI Act creates specific compliance requirements for AI used in media. Some applications (Content moderation, recommendation systems, deepfake detection) may qualify as high-risk under Annex III, requiring conformity assessment and ongoing monitoring from 2 August 2026.

Updated June 2026 · MmowW AI Compliance

EU AI Act Impact on Media: What Changes in August 2026

EU AI Act and Media

The EU AI Act affects media through multiple pathways. Certain AI applications in this sector — particularly content moderation, recommendation systems, deepfake detection — may qualify as high-risk under Annex III, triggering the full suite of compliance obligations from 2 August 2026.

Even for non-high-risk AI applications, Article 4 AI literacy obligations apply to all staff using AI tools. This means documenting your AI use, ensuring staff competence, and maintaining records of AI-related decisions.

High-Risk Classification in This Sector

In media, AI applications most likely to qualify as high-risk include those that: make or significantly influence decisions affecting individuals, process sensitive personal data, operate as safety components of regulated products, or affect access to essential services.

Conduct a thorough review of your AI applications against Annex III categories. Document your classification reasoning — this becomes part of your compliance evidence.

Practical Compliance Steps

1. AI Inventory: List all AI tools and their purposes in your operations.

2. Risk Classification: Identify which applications in media may qualify as high-risk.

3. Documentation: Prepare technical documentation for high-risk systems per Article 11.

4. Human Oversight: Ensure competent human oversight for high-risk AI decisions per Article 14.

5. Monitoring: Establish post-deployment monitoring for high-risk systems per Article 61.

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.