Prohibited practices & AI literacy
Certain AI practices are banned; companies must ensure sufficient AI literacy among their staff.
Build AI literacy through training and rule out prohibited applications.
AI Governance · EU AI Act · KI-MIG
Make AI responsibly usable
A lean framework suited to mid-sized companies that creates order – with a policy, roles, risk logic and an inventory of your AI applications. With an eye on the transparency obligations from August 2026.
EU AI Act & KI-MIG
What you need to know now
The EU AI Act applies EU-wide and in stages; Germany's AI Market Surveillance and Innovation Promotion Act (KI-MIG) implements it nationally – designating the competent supervisory/market-surveillance authority and detailing enforcement and fines. The key deadlines – and what to do:
Governance & penalties
Obligations for general-purpose AI models (GPAI, e.g. the models behind ChatGPT), supervisory structures and penalty rules take effect.
Clarify responsibilities and list where AI is used across the company (AI inventory).
Transparency obligations
AI systems such as chatbots must be recognisable as AI; AI-generated content must be labelled.
Label chatbots & AI content, prepare user notices.
High-risk AI
Extended obligations for high-risk systems, e.g. in regulated products (longer transition period).
Check early whether your AI qualifies as high-risk – and plan for conformity.
KI-MIG is still in the legislative process; details may yet change. We keep track of the status for you. This is not legal advice.
Note: We do not provide legal advice. Our governance and EU AI Act services are structured professional orientation and implementation preparation, subject to legal review in each case.
Why now
Many employees already use AI on their own – without rules and without regard for data protection and risk ("shadow AI"). The EU AI Act brings staggered obligations; from 2 August 2026, for example, chatbots must be clearly recognisable as AI. Creating order now avoids scramble and risk later.
The governance framework in detail
In concrete terms you get five coordinated building blocks – a framework you actually apply day to day, not a paper tiger for the shelf.
- An overview of where AI is used across the company (AI inventory)
- Clear usage rules: what is allowed, what is not, which data may go in, what must be labelled
- Who is responsible for what (role model)
- Classification by risk, aligned with the EU AI Act
- An approval/review process for new AI plus a short awareness briefing for the team
EU AI Act quick check
For acute need to act: a focused risk classification of your AI applications plus a prioritised list of measures – in 1–2 weeks. You then know what to do by when.
Matching packages
How we can work together
Governance On request
AI Governance Starter
A lean, practical framework: policy, roles, risk logic, AI inventory – with an eye on the EU AI Act.
3–5 weeks Learn moreGovernance On request
EU AI Act Quick Check
A structured risk classification of your AI applications plus prioritised actions – time-critical from August 2026.
1–2 weeks Learn moreEnablement On request
AI Training Programme
Hands-on, role-specific enablement – from ChatGPT basics and safe tool use to the EU AI Act.
modular, 0.5–6 days Learn moreFAQ
Frequently asked questions
Is this a legally binding review?
No. We provide structured technical orientation. The legally binding assessment is made by your legal advisor. We prepare clean input for them.
What happens from 2 August 2026?
Among other things, transparency obligations take effect: AI systems such as chatbots must be recognisable as AI, and AI-generated content must be labelled. The quick check shows your specific need to act.
How EU AI Act-ready is your AI use?
The quick check classifies your applications and shows what needs to be done by when.