Submitted text
Artificial intelligence (AI) is rapidly playing an increasingly important role in our society. The potential for progress offered by these technologies is immense. So are the challenges they bring with them.
AI has the potential to radically change the way we work, communicate and even live together. If it is used incorrectly or for the wrong purposes, it can also have serious negative consequences, particularly for privacy, security or human dignity.
In this context, I would like to put the following questions to the Federal Council:
- Is he ready to launch a national research program that looks at how AI will affect our society and democracy, what framework conditions can be created for it and how it can be developed for purposes that benefit everyone?
- Does it plan to create a specific legal framework for technologies that use AI, or does it consider the current legislation to be sufficient?
- How does he think the new EU regulation will affect Swiss legislation? Will Switzerland have to take measures to adapt or to obtain recognition of equivalences?
- What impact can the ongoing negotiations on a Council of Europe Convention on “Artificial Intelligence, Human Rights, Democracy and the Rule of Law” have on Switzerland and its legislation? The preliminary draft of the convention was recently published by the Committee on Artificial Intelligence (CAI), which is chaired by a Swiss national. Is Switzerland planning to adapt its legislation? Has work already begun in this area?
- What measures should be taken to ensure that the development of these technologies finds a regulated but favorable framework in Switzerland and that companies active in this field are founded or set up in Switzerland?
- Shouldn’t Switzerland develop a comprehensive strategy that covers these different aspects? Is the Federal Council prepared to tackle this task?
Opinion of the Federal Council of 30.8.2023
The Federal Council has already taken numerous measures to ensure the responsible use of artificial intelligence (AI). On the basis of the report published in 2019 by the interdepartmental working group on artificial intelligence, it has drawn up guidelines for the use of AI in the Federal Administration and has published the Competence Network for Artificial Intelligence (CNAI) was launched.
There are established procedures for launching a national research program. Interested parties can submit topic proposals according to the bottom-up principle. Based on these, the State Secretariat for Education, Research and Innovation (SERI) draws up program proposals, which are checked for feasibility by the Swiss National Science Foundation. The Federal Council has already launched the National Research Program NRP 77 “Digital Transformation”, which also includes research on AI – particularly in the research module “Ethics, Trustworthiness and Governance”. A proposal on digital democracy was also examined in the NRP 2022/2023 review round. This review was negative, partly because many overlaps with NRP 77 were identified (see responses to Inquiry 23.7210 and Interpellation 23.3281).
The effects of the use of AI are highly context-dependent. For this reason, the Federal Council has already taken regulatory measures for certain areas. In particular, it has instructed the Federal Department of the Environment, Transport, Energy and Communications (DETEC) to prepare a draft on the regulation of large communication platforms. This draft is intended to strengthen the rights of users in Switzerland and increase the transparency requirements for platforms. The new Data Protection Act, which comes into force on September 1, 2023, also aims to adapt data protection to the state of the art. It contains provisions on decisions based exclusively on automated data processing. As part of the current consultation on the amendment to the Copyright Act (ancillary copyright for media publishers), it is to be determined whether there is a need for regulation with regard to the use of generative AI.
At an international level, the Federal Council is following the internal EU negotiations on the regulation establishing harmonized rules for artificial intelligence (Artificial Intelligence Act, “AI Act”). These negotiations are expected to be concluded at the end of 2023 or beginning of 2024. Before a possible adaptation of the Swiss legal framework to regulate the use of AI in line with the AI Act is considered, the consequences must be analyzed in detail and the interests carefully weighed up. However, such an analysis can only be carried out once a final regulation text is available. In addition, Switzerland is actively involved in the Council of Europe’s negotiations on a convention in this area, contributing its interests and values. Switzerland is chairing the relevant committee. If Switzerland ratifies the Convention, it will then be obliged to implement it nationally. Whether legal adjustments are necessary will only become clear once the negotiations have been concluded.
In its statement on the Postulate Dobler (23.3201) the Federal Council has already stated that it will carry out a political analysis within the framework of existing federal bodies (in particular the interdepartmental EU Digital Policy Coordination Group, the Tripartite Platform, the AI Guidelines Monitoring and the FSO’s AI Competence Network (CNAI)). All federal agencies responsible for the legal areas concerned are involved in this work. The need for action and possible options for sectoral and, if necessary, horizontal measures will be identified by the end of 2024.