Guidelines on Artificial Intelligence for the Confederation

The ‘Guidelines on Artificial Intelligence for the Confederation’ document issued by the Federal Council outlines strategic guidelines for addressing the challenges of AI at the federal level in Switzerland. Below are the key guidelines mentioned in the document:

1. Putting People First

• Dignity and Well-being: AI development and usage should prioritize human dignity and well-being, ensuring self-determination and participation in social and political life.
• Improving Quality of Life: AI should enhance the quality of life and promote sustainable societal development.
• Ethical Considerations: Fundamental rights and ethics must be integrated into AI design and usage, with measures to prevent discrimination and protect privacy.

2. Regulatory Conditions for AI

• Innovation-Friendly Regulations: The Confederation aims to create favorable regulatory conditions to foster AI innovation and maintain Switzerland’s competitive edge in AI research and development.
• Balanced Regulation: A balanced approach is required to encourage economic growth and sustainable development through AI while maintaining legal certainty.

3. Transparency, Traceability, and Explainability

• Trustworthy AI: Transparency, traceability, and explainability are crucial for AI systems to be trustworthy.
• Data Disclosure: The functioning and data sets used in AI should be disclosed as per legal obligations to enable monitoring and compliance.

4. Accountability

• Clear Liability: Responsibility for AI decisions and outcomes should be clearly defined, ensuring that liability cannot be delegated to machines.

5. Safety

• Robust Design: AI systems must be designed to be safe, robust, and resilient, with safeguards to prevent misuse and mitigate risks to individuals, society, and the environment.

6. Actively Shape AI Governance

• International Involvement: Switzerland should actively participate in global AI governance and contribute to the development of international standards and norms.

7. Involve All Relevant Stakeholders

• Inclusive Decision-Making: AI governance should include national and international stakeholders from various sectors, ensuring accountability and effective cooperation.

Monitoring and Adaptation

• Continuous Monitoring: Advances in AI should be closely monitored, and the guidelines should be regularly assessed and updated in alignment with the Digital Switzerland Strategy and the Digital Foreign Policy Strategy for 2021-2024.

Specific Guidelines for Education, Research, and Innovation

• Digital Skills: The education system must ensure that citizens acquire basic digital skills and advanced AI training.
• Societal Impact: Research and education should focus on the societal and environmental impacts of AI, promoting transparency and ethical considerations.

Applicability of Legal Standards

• Existing Legal Framework: AI development and usage are subject to existing international and national laws, including fundamental rights, intellectual property, liability, and non-discrimination laws.
• New Data Protection Act: The revised Data Protection Act includes provisions specifically relevant to AI, such as profiling, genetic and biometric data, privacy by design, impact assessments, and automated decision-making.

These guidelines aim to ensure the responsible development and application of AI within the Federal Administration, promoting innovation while addressing the associated risks and ethical considerations.