MSC Welcomes the “Frontier AI Safety Commitments” and Urges Attention to the Impact of AI on Worldwide Elections

On May 21, 2024, the governments of the United Kingdom and the Republic of Korea have announced the agreement of multiple international technology companies to a set of commitments intended to ensure the safe and responsible development of highly capable general-purpose AI models. The “Frontier AI Safety Commitments” were presented at the AI Seoul Summit and include the transparent measurement and mitigation of risks associated with AI models and systems. Companies from the United States, Europe, the Middle East, and Asia were among the 16 that agreed to the commitments.

Ambassador Christoph Heusgen, Chairman of the MSC, commented, “As the convener of the ‘Tech Accord to Combat Deceptive Use of AI in 2024 Elections,’ the MSC welcomes this initiative by the UK and South Korean governments and the agreement between companies, including signatories of the Accord. It is important that our signatories continue to promote cooperation on AI governance wherever they can.” Heusgen added, “In particular, we commend the commitment to transparency and accountability. As this technology advances rapidly, these practices are critical to enable sustainable and democratic governance of AI.” As stated by the AI Elections Accord, transparency should not only guide the research, development, and implementation of AI models but also the active efforts to mitigate the potentially harmful impact of AI-generated content on public discourse.

In response to the impact of AI-generated content on electoral processes, 23 leading technology companies pledged to help prevent deceptive AI content from interfering with this year’s global elections by signing the “Tech Accord to Combat Deceptive Use of AI in 2024 Elections” at the last Munich Security Conference in February 2024. This timely initiative stood at the beginning of a year in which more than four billion people in over 40 countries were invited to vote – more than ever before in human history. Considering major upcoming elections, including elections in the European Union, the United Kingdom, and the United States, the commitments of the AI Elections Accord gain increased urgency and importance. The “Frontier AI Safety Commitments” are a welcome complement to the pledges made in the AI Elections Accord in Munich and should be used for momentum to intensify these efforts.

The AI Elections Accord is a set of commitments to deploy technology countering harmful AI-generated content meant to deceive voters. Signatories pledge to work collaboratively on tools to detect and address online distribution of such AI content, promote educational campaigns, and provide transparency, among other steps. It also includes a broad set of principles, including the importance of tracking the origin of deceptive election-related content and the need to raise public awareness about the problem.

In its capacity as the convener of the AI Elections Accord, the Munich Security Conference supports the Accord’s signatories and organizes selected thematically related in-person events throughout the year as well as at the margins of the main conference in February 2025. More information about the AI Elections Accord can be found on our website.

Please reach out to us via press@securityconference.org if you would like to have more information about the initiative.