Artificial intelligence is reshaping industries, economies, and daily life—but who sets the rules? In Switzerland, the Federal Council is stepping up to answer that question. While the country has no overarching AI-specific laws yet, the government is charting a path that balances innovation with public trust and fundamental rights. On February 12, 2025, the Federal Council decided to ratify the AI Convention and published a report outlining its approach to AI regulation. The report primarily focuses on the Council of Europe's AI Convention while also considering potential alignment with the EU’s AI Act.

Key Takeaways from Switzerland’s AI Regulation Report

1. A Gradual and Sector-Specific Approach
   

   • Switzerland aims to regulate AI through sector-specific legislative changes, with a focus on areas where AI poses the highest risks and opportunities, as identified in the Federal Council's analyses. rather than a single overarching AI law.

   • Fundamental rights-related areas like data protection may be addressed through cross-sectoral rules.

   
   

2. Adopting the AI Convention with a Swiss Twist
   

   • Switzerland plans to ratify the Council of Europe's AI Convention, ensuring AI use aligns with democratic principles and human rights. The Convention's obligations will be more extensive for state actors than for private entities, with specific regulatory requirements primarily affecting public institutions.

   • The Convention will primarily apply to state actors, with initial guidelines already outlined for the private sector, though further details will be determined through follow-up measures. However, private sector self-regulation alone will not be sufficient, and government measures will likely be required to ensure compliance with fundamental rights and transparency standards.

     
     

3. Aligning with the EU AI Act?
   

   • The Federal Council is evaluating the potential alignment of Swiss law with the EU AI Act, recognizing that full alignment would introduce significant regulatory obligations. However, there is ongoing debate over whether Switzerland should pursue a selective approach rather than full adoption. acknowledging that full alignment would impose significant regulatory obligations on Swiss businesses, particularly in terms of compliance with risk-based AI classifications and market access requirements.

   • Full alignment could facilitate market access for Swiss AI companies, though it presents a trade-off between regulatory obligations and economic benefits, with ongoing discussions on whether a more selective approach should be pursued.

     
     

4. Transparency, Trust, and Oversight
   

   • A draft bill by the end of 2026 will focus on transparency, non-discrimination, and data protection, with additional oversight mechanisms and risk assessments required under the AI Convention to ensure compliance and mitigate potential harms.

   • The government is exploring non-binding measures such as self-disclosure agreements and industry-led compliance solutions.

     
     

5. Three Regulatory Approaches Considered
   

   • Continuation of sector-specific regulatory activities: AI regulation remains fragmented across industries, addressing risks as they arise.

   • Ratification of the AI Convention with minimal or extensive implementation: A structured but flexible approach that prioritizes transparency and human rights.

   • Alignment with the EU AI Act: Would impose stricter compliance but enable smoother integration into the European market.

Economic and Trade Considerations

• The EU AI Act impacts Swiss companies exporting to the EU, necessitating compliance with European standards.

• Switzerland's Mutual Recognition Agreement (MRA) with the EU plays a critical role in ensuring that Swiss products with AI components remain competitive in the EU market. However, its extension to AI-specific regulations is uncertain and depends on broader Swiss-EU negotiations.

• Swiss companies are cautious about adopting the EU AI Act wholesale, citing concerns over bureaucracy and innovation restrictions.

Federal Council’s Official Position on AI Regulation

• The AI Convention of the Council of Europe will be incorporated into Swiss law, applying primarily to state actors.

• Legislative changes will focus on sector-specific adaptations, with only fundamental rights-related areas subject to broader regulation.

• Non-binding measures, such as self-disclosure agreements or industry-led compliance frameworks, will complement legal reforms.

• AI regulation will be structured around three core objectives: reinforcing Switzerland as an innovation hub, safeguarding fundamental rights (including economic freedom), and enhancing public trust in AI.

Next Steps in Swiss AI Regulation

• The Federal Department of Justice and Police (FDJP) will draft a bill in consultation with DETEC and the FDFA by the end of 2026.

• The bill will focus on transparency, data protection, non-discrimination, and supervision to implement the AI Convention effectively.

• DETEC, FDJP, FDFA, and EAER will also develop a roadmap for non-legally binding measures, ensuring Switzerland’s approach aligns with its key trading partners. This roadmap will include industry self-regulation, voluntary disclosure agreements, and other flexible mechanisms to encourage compliance while maintaining regulatory agility.

• Stakeholder consultations, involving both internal and external experts, will shape the final regulatory framework.

Balancing Innovation and Regulation

Switzerland’s approach seeks to reinforce its position as an innovation hub while ensuring AI operates within a legal framework that protects fundamental rights. The Federal Council’s decision reflects a pragmatic approach: leveraging existing international frameworks while tailoring regulations to Switzerland’s needs.

With the AI Convention ratification on the horizon and a draft bill expected in 2026, the regulatory landscape in Switzerland is set to evolve rapidly. Staying ahead of these changes will be crucial for businesses leveraging AI technologies.