From August 2026, German companies must systematically train their employees in AI competency. This is mandated by the national implementation of EU Regulation 2026/1744, known as the Digital Omnibus. For the first time, lawmakers are establishing binding requirements for how organizations handle artificial intelligence in daily operations – rules that apply to both management and staff.
Key Facts
- Effective date: August 2026 – AI training obligation for all German companies
- Supervising authority: The Federal Network Agency (Bundesnetzagentur) has overseen central market surveillance since August 2, 2026
- Penalties: Up to €35 million or 7% of global annual turnover for prohibited AI practices; up to €15 million or 3% of turnover for violations involving high-risk systems and transparency requirements
- Transition period: Generative AI systems have until December 2, 2026 to technically implement labeling systems
What the Competency Requirement Entails
The objective is straightforward: both managers and employees must develop a basic understanding of how AI systems function and their associated risks. The regulation builds on developments initiated as early as February 2025.
One important caveat: while competency promotion remains mandatory, the current version of Article 4 of the EU AI Regulation (amended July 27, 2026) contains no direct fines for merely failing to provide training. However, experts caution that competency is a prerequisite for complying with other sanctioned obligations – companies that skip staff training therefore indirectly violate other rules and expose themselves to penalties.
Transparency Requirements and Labeling Obligations
Alongside the competency mandate, transparency requirements under Article 50 take full effect in August 2026. This means:
| Requirement | Deadline | Applies to |
|---|---|---|
| Chatbot labeling | August 2026 | All AI applications |
| AI-generated content labeling | August 2026 | All AI applications |
| Technical implementation (generative AI) | December 2, 2026 | Generative systems |
Companies must not only train their teams but also ensure their AI systems are transparently labeled – failure to do so triggers fines.
The Penalty Framework: Tiered by Severity
The EU AI Regulation establishes a differentiated penalty system reflecting violation severity:
- Prohibited AI practices: Up to €35 million or 7% of global annual turnover
- High-risk systems & transparency violations: Up to €15 million or 3% of turnover
- Misleading information to authorities: Up to €7.5 million or 1% of turnover
For context: nine specific AI practices have been banned since February 2025. In December 2026, another ban takes effect – the creation of non-consensual intimate imagery. Requirements for high-risk systems are phased in over time.
What This Means for Your Organization
The AI competency requirement is not toothless – it forms the foundation for compliance with all other regulations. Companies should begin developing and documenting training programs now. The Federal Network Agency will audit compliance starting in August. Organizations that delay until regulators arrive risk not only fines but also reputational damage. Particularly mid-market firms should engage with this issue early – the timeline to August 2026 is tight.
Sources
Editorially owned by Ideal Syka. Sources and method: Newsroom & method. Tips and corrections: ai@i6eal.de.
