EC Seeks Input on AI Misinfo Guidelines for Elections

The European Commission has taken steps to protect upcoming European elections from misinformation by introducing guidelines that require tech platforms like TikTok, Facebook, and X to detect artificial intelligence (AI)-generated content. The recommendations are part of a public consultation on election security guidelines for very large online platforms (VLOPs) and very large online search engines (VLOSEs). The goal is to address the democratic threats posed by generative AI and deepfakes. Generative AI can create and spread misleading synthetic content about political actors, events, election polls, and narratives, which can mislead voters and manipulate the electoral process.

The draft guidelines are currently open for public consultation in the European Union until March 7. They propose measures such as alerting users on relevant platforms to potential inaccuracies in content produced by generative AI and guiding users to authoritative information sources. The guidelines also recommend that tech giants implement safeguards to prevent the generation of misleading content that could significantly influence user behavior. For AI-generated text, the draft suggests that platforms should indicate the sources of information used as input data to enable users to verify the reliability of the content.

The draft guidance draws inspiration from the recently approved AI Act in the European Union and its non-binding counterpart, the AI Pact. Concerns about advanced AI systems, like large language models, have grown since the viral surge of generative AI in 2023. The commission has not specified when companies will be required to label manipulated content under the EU’s content moderation law, the Digital Services Act. Meta, the parent company of Facebook, has announced plans to introduce fresh guidelines for AI-generated content on its platforms in the coming months. Any content identified as AI-generated will be labeled accordingly, whether through metadata or intentional watermarking.

The European Commission’s efforts to address the threat of AI-generated content in elections aim to safeguard the democratic process and reduce the potential spread of misinformation. By requiring tech platforms to implement measures to detect and label such content, the commission hopes to prevent the manipulation of voters and ensure that users have access to reliable and accurate information. The public consultation on the draft guidelines allows for input from various stakeholders, and the final guidelines will provide clear guidance for European Parliament elections. This initiative aligns with the EU’s broader commitment to regulating AI technology and ensuring its responsible and ethical use in society.