HR 5272
Protect Elections from Deceptive AI Act
May require changes to AI practices. Monitor and prepare.
TL;DR
Rep. Julie Johnson (D-TX) introduced this bill to ban AI-generated deepfakes that falsely depict federal candidates in campaign ads and political communications. It would let candidates sue creators and distributors of deceptive AI content, and give the FEC authority to enforce violations. Think fake Biden robocalls or fabricated Trump videos during election season.
How This Might Impact Your Business
Political advertising platforms, including Meta, Google, X, and TikTok, would face legal exposure for distributing AI-generated deepfakes of federal candidates
Generative AI companies (OpenAI, Anthropic, Midjourney, ElevenLabs) could see new pressure to add watermarking, provenance tracking, and political content filters to their tools
Political consultants and ad agencies using AI-generated voiceovers, images, or video of real candidates would need explicit consent or clear satire/parody labeling
Broadcasters and streaming services running political ads would need verification processes to confirm AI content authenticity before airing
Candidates depicted in deceptive AI content could sue for injunctive relief and damages, creating direct litigation risk for content creators and distributors
Exemptions likely apply to bona fide news coverage, satire, and parody, though the line between satire and deception will be litigated
No specific revenue threshold or company size carve-out, meaning small AI startups face the same rules as large platforms
What Should You Do
Ask your legal team whether your AI tools or ad platforms could be used to generate or distribute deepfakes of federal candidates, and document your content moderation policies
If you run a generative AI product, accelerate watermarking and content provenance features (C2PA standards) before the 2026 election cycle
Political ad buyers and agencies should require written attestations from creative vendors confirming no unauthorized AI likenesses of candidates
Monitor the House Administration Committee for hearings and markup; pair this with state-level deepfake laws already active in California, Texas, and Michigan
Review your terms of service and acceptable use policies to explicitly prohibit election-related deepfakes if they don't already
Who It Affects
Sponsors
Status Timeline
committee
Referred to the House Committee on House Administration.
September 10, 2025
AI-generated analysis for informational purposes only. Not legal advice. Always consult a qualified attorney for legal guidance.
Need help preparing your team for AI compliance?
Talk to LaunchReady about AI Training