Human responsibility
AI may prepare decisions, but must not obscure responsibility. Critical decisions need clear accountability and human control.
Responsible use of AI
Ethics & AI governance for fair, sovereign AI.
AI should empower people, improve work in meaningful ways and remain accountable - not surveil, manipulate or hide responsibility.
It matters to me that artificial intelligence is used ethically, morally responsibly and fairly. That is why I see governance not as bureaucracy, but as the foundation for trust, data sovereignty and sustainable systems.
Stance
AI may increase productivity, but must not undermine dignity, fairness or self-determination.
Good AI architecture combines value, data protection, security and clear responsibility.
Rules must be understandable enough for teams to actually apply them day to day.
Core principles
Ethical AI needs clear guardrails.
AI governance begins before buying any tool: with purpose, responsibility, data flows, risk assessment and the question of what impact a system has on people.
Fairness and non-discrimination
AI systems must be checked for bias, exclusion and unfair effects - especially with sensitive data, assessments and recommendations.
Data protection and data sovereignty
Personal, confidential and business-critical data belongs under the control of the organisation - technically, organisationally and contractually.
Traceability
Good AI governance documents purpose, data flows, risks, limits and responsibilities so that decisions remain explainable later on.
What I support
AI as a tool for better work.
- AI that relieves specialists and prepares better decisions
- Automation that reduces repetitive work without outsourcing responsibility
- Sovereign AI architectures with EU hosting, on-premise options and clear data flows
- Transparent systems with documented limits, risks and responsibilities
- Practical governance that enables teams instead of blocking them
What I do not support
No AI against people.
- Mass surveillance, social scoring or covert behavioural control
- Weapons, autonomous use of force or systems for escalating conflict
- AI as a pretext to replace staff wholesale instead of improving work meaningfully
- Manipulative systems, dark patterns or opaque decision logic
- Uncontrolled use of sensitive data in external tools without clear legal and risk review
AI governance
Lean rules instead of uncontrolled tool sprawl.
Governance has to be practical. It should give teams orientation, reduce risk and document why an AI system is used - and where its limits lie.
1. Clarify purpose
Which problem should AI solve, which decision does it support, and where must it explicitly not be used?
2. Assess risks
Data protection, fairness, security risks, faulty domain decisions and possible impacts on people are reviewed before implementation.
3. Secure the architecture
Data flows, models, hosting, access concepts, logging and integrations are planned so that control is preserved.
4. Document operations
Roles, responsibilities, approvals, evaluations and limits are recorded transparently and reviewed regularly.
Frequently asked questions
Questions about ethical AI and governance.
What does ethical AI mean in practice?
Ethical AI means that the purpose, data, risks, responsibilities and effects of a system are deliberately reviewed. The goal is not maximum automation at any cost, but fair, safe and traceable use.
Is AI governance only a topic for large companies?
No. Smaller and medium-sized companies also need clear rules when teams use AI with customer data, internal knowledge or operational processes. Governance can be lean, but it has to be traceable.
Does AI governance replace legal advice?
No. AI governance creates technical and organisational clarity, but does not replace a legal review. For GDPR, the EU AI Act or employment-law questions, legal advice should be involved.
Next step
Introduce AI responsibly.
If you want to use AI without sacrificing control, fairness or data protection, a short governance check before technical implementation is worth it.