As voters across Europe prepared for the 2024 European Parliament elections, a new threat emerged on the political horizon: AI-generated deepfakes and manipulated content. While deepfakes had yet to make a meaningful impact on elections, the technology industry, political parties, and civil society organizations recognized the urgent need to prepare defenses to protect democratic processes.
A Collective Industry Response
In February 2024, 27 leading technology companies signed the Tech Accord to Combat Deceptive Use of AI in 2024 Elections at the Munich Security Conference. This unprecedented collective effort represented a unified commitment to address the challenges posed by AI-manipulated content in electoral contexts.
The accord focused on several key areas:
- Developing tools like Content Credentials to detect and address online distribution of deepfakes
- Advancing content provenance and watermarking technologies
- Working toward a common approach such as the C2PA standard
- Driving public awareness through education campaigns
- Providing transparency in AI-generated content
Microsoft took numerous additional steps to fulfill these commitments and launched new tools to stay ahead of evolving threats against voters, candidates, political campaigns, and institutions across Europe.
Supporting Political Parties and Campaigns
Political parties do not want their campaigns hijacked by disinformation or voters misled by manipulated content. Microsoft focused much of its European efforts on supporting them through:
Awareness and Education
Microsoft organized awareness sessions in Brussels and across all 27 EU Member States with political staffers and candidates. These sessions provided information on deepfake risks and solutions to protect campaigns and react effectively.Content Integrity Tools
Microsoft expanded its Content Integrity tools to all EU campaigns and news organizations, enabling them to trace media origin and combat deepfakes. Parties could add secure "Content Credentials" to their content, providing information about who created it, where and when it was made, and whether it had been edited. These credentials also made it easier for media to fact-check content.Reporting Mechanisms
Microsoft created a public website where users could report deceptive AI-generated media appearing on its consumer services.Protecting Democratic Institutions
Recognizing that political campaigns, journalists, think tanks, human rights organizations, and non-profits face attacks from bad actors, Microsoft offered additional protection layers:
Microsoft AccountGuard
This cybersecurity protection service was made available to organizations supporting healthy democracies in 23 EU Member States and in the European Parliament.Election Communications Hub
Microsoft established a direct connection between EU governments, election authorities, and Microsoft security teams to address security challenges during and leading up to elections.Raising Public Awareness
Empowering the public through education represents the first line of defense against deepfakes. Microsoft launched the "Check. Recheck. Vote." campaign across all 27 EU Member States, aimed at:
- Educating EU citizens about deepfakes
- Promoting critical evaluation of voter information
- Highlighting the role everyone plays in curbing disinformation spread
- Promoting trusted official EU sources for voting information
Societal Resilience Fund
Together with OpenAI, Microsoft launched the Societal Resilience Fund to provide grants promoting AI education and training among voters and vulnerable communities.Supporting Official Campaigns
Microsoft supported the European Parliament's "Use Your Vote" campaign on Bing Search by including search results tailored to all 24 official EU languages with links to official European Parliament sources and dedicated banners, ensuring citizens had accurate information on when, where, and how to vote.EU Code of Practice on Disinformation
As a signatory to the EU Code of Practice on Disinformation, Microsoft introduced specific measures related to the European elections, including a rapid response mechanism, and regularly reported on its work in the Code's Transparency Centre.
Long-term Commitment to Democracy
Microsoft's commitment to defending democracy extends beyond a single election cycle:
- 2016: Established the Democracy Forward team in response to emerging cyber threats
- 2020: Expanded focus to include threats against democratic institutions and the information environment
- 2023: Made specific commitments to protect election integrity with AI-focused measures
The Ongoing Challenge
While large-scale AI-based manipulation may not have materialized during the 2024 European elections, the threat remains real for future electoral cycles. Vigilance and continued innovation in defense mechanisms are essential.
"We all need to stay vigilant and do our part to avoid AI being misused to chip away at democratic processes," Microsoft emphasized. "We will continue to build defenses and work together with society, governments, and industry, to protect the vitality and longevity of our democracies."
The fight to protect democratic integrity in the age of AI requires sustained collaboration among technology companies, governments, civil society, and individual citizens—each playing a vital role in ensuring that elections remain free, fair, and trustworthy.
Source: Microsoft EU Policy Blog