Innovation-Friendly Adoption of AI Convention Rather Than a «Swiss AI Act»

The Swiss Federal Council decided on February 12, 2025 that it does not support the implementation of broad, horizontal AI regulation in Switzerland. Thus, there will be no proposal for a «Swiss AI Act» or similar equivalent to the EU’s comprehensive AI Act (see our Bulletin of July 16, 2024 for details on the EU AI Act).

Instead, the Swiss Federal Council prefers to implement the Council of Europe’s AI Convention by making limited sector-specific adjustments to existing laws and limiting general, cross-sector regulation to key areas relevant to fundamental rights, such as data protection. According to the Swiss Federal Council, the Swiss regulatory approach to AI shall be based on three objectives: strengthening Switzerland as a location for innovation, ensuring the protection of fundamental rights, including economic freedom, and increasing public trust in AI. Another important focus is to ensure the compatibility of the Swiss regulatory approach with the regulations of its main trading partners, in particular the EU. The aim is to avoid any «Swiss finish», i.e., additional regulatory burdens on Swiss companies compared to market players in the EU.

Outcome of the Review of the AI Regulatory Regime in Switzerland

The decision of the Swiss Federal Council is based on an in-depth analysis commissioned by the Federal Council on November 22, 2023 (see our Bulletin of November 23, 2023). The objective of the analysis was to identify the need for regulatory action and to provide a basis for decision-making regarding the future shape of Swiss AI regulation. It was published on February 12, 2025.

A particular focus of the analysis was to examine the potential impact of ratifying the AI Convention. According to the analysis, Swiss law already contains legal norms applicable to AI, which would allow a partial adoption of the AI Convention. However, for a full adoption, additions are necessary, in particular to improve the transparency of AI systems, to assess their impact on fundamental rights and to ensure control mechanisms. The Swiss Federal Council has decided to prioritize sector-specific regulations and to introduce cross-sectoral regulations only where necessary.

What is the AI Convention of the Council of Europe?

The AI Convention is an international framework agreement to ensure that AI systems respect human rights, democracy and the rule of law throughout their life cycle, from design to decommissioning. It requires ratifying states to adopt legal and administrative measures to ensure ethical, transparent and non-discriminatory AI. At the same time, it promotes innovation under safe conditions and establishes international cooperation and monitoring mechanisms. The Convention does not regulate AI as such; it only addresses aspects of AI that relate to the protected interests mentioned above. It is also important to note that the AI Convention itself does not prescribe specific prohibitions, but requires ratifying states to assess whether specific regulation is necessary.

What Are the Next Steps?

The Swiss Federal Council has asked the federal administration, led by the Federal Department of Justice and Police to draft a proposal to incorporate the AI Convention into Swiss law. The draft should be available for public consultation by the end of 2026.

In addition to legislative measures, non-binding legal measures will be developed. These include self-declaration agreements and industry-specific solutions that provide greater flexibility for companies.

Which Regulatory Changes Are to be Anticipated?

The Federal Council has not yet decided which specific regulatory changes it will submit to parliament. However, it is generally expected that the proposal will include stricter transparency requirements, mandatory labeling of AI-generated content, and regulations governing political advertising on social media. In addition, data protection law may be revised, in particular to prevent AI-driven surveillance, such as facial recognition, or with regard to information obligations. In addition, the obligation to conduct a privacy impact assessment may be extended, with differentiated obligations for different stakeholders. Equality and non-discrimination issues are also expected to be addressed. Targeted rules are expected to be proposed for key sectors, including finance, healthcare and digital platforms.

What is the Impact on Swiss Businesses?

The Federal Council’s strategy aims to strengthen the regulation of AI risks without imposing excessive regulatory burdens on Swiss companies. This innovation-friendly strategy can enable flexible solutions for companies and strengthen Switzerland’s position as a leading hub for AI development and applications. It remains to be seen, however, what specific legislative changes will be proposed and to what extent such proposals will become law in Switzerland.