On February 16, 2024, at the Munich Security Conference, twenty technology companies announced “A Tech Accord to Combat Deceptive Use of AI in 2024 Elections.” The signatories were Adobe, Amazon, Anthropic, Arm, ElevenLabs, Google, IBM, Inflection AI, LinkedIn, McAfee, Meta, Microsoft, Nota, OpenAI, Snap, Stability AI, TikTok, TrendMicro, Truepic, and X. The accord was framed around a year in which more than four billion people across over 40 countries were due to vote.
The accord targets AI-generated audio, video, and images that deceptively fake or alter the appearance, voice, or actions of political candidates and election officials, or that give voters false information about when, where, and how to vote. The companies committed to a set of goals: developing and deploying technology to mitigate these risks, evaluating their own models, working to detect and address such content on their platforms, attaching provenance signals where feasible, building cross-industry resilience, providing public transparency, engaging civil society and academics, and supporting public awareness and media-literacy efforts.
The commitments are voluntary and aspirational rather than legally binding, and critics noted that the language (“reasonable precautions,” “where appropriate”) left enforcement to each company. It was nonetheless the broadest single coordination point among rival AI labs, social platforms, and content tools on election integrity.
For a business reader, the accord is a marker of how the industry tried to self-regulate around a high-stakes risk before governments acted, and a reference list of which provenance and detection practices the major players publicly endorsed.