A bipartisan duo is calling on federal agencies to coordinate efforts to thwart potential election threats from artificial intelligence, such as chatbots giving voters inaccurate information.
Reps. Mike Lawler, New York Republican, and Josh Gottheimer, New Jersey Democrat, addressed the heads of the Cybersecurity and Infrastructure Security Agency, Justice Department, Federal Election Commission and Department of Homeland Security in their Tuesday letter.
They urged the four agencies to prepare for AI-driven threats to election integrity, such as deploying monitoring and mitigation capabilities, and “deepen engagement” with major AI companies regarding safeguard protocols, testing practices and how political bias, misinformation or model vulnerabilities are identified.
“As millions of Americans rely on AI-driven tools to research the upcoming election, the accuracy and neutrality of these tools are now directly tied to the integrity of our democratic process,” the two House members wrote.
In May, the lawmakers wrote to major AI companies — OpenAI, Google, Microsoft, Anthropic, Meta, Perplexity and X — concerning the role AI will play in November’s midterm elections and their responsibility in operating such platforms.
In both letters, Mr. Lawler and Mr. Gottheimer also cited concerns that AI models use opinion websites, such as Reddit and Wikipedia, that present unverified information.
“By relying on this information, systems cannot reasonably present a full picture of unbiased information to voters,” they wrote.
They said that in the companies’ responses, they “acknowledged both the progress they made and the limits of voluntary safeguards in addressing political bias and misinformation.”
The AI companies broadly converged on neutrality-by-design and sourcing and citation transparency, while the model developers among them also pointed to shared content-provenance standards. None offered binding new commitments.
“Their responses made clear that, while industry can improve transparency and curb certain abuses, the scale and speed of AI-driven information demands a more coordinated public response,” Mr. Lawler and Mr. Gottheimer wrote.
Ahead of this year’s midterms, AI-driven methods for receiving information “must not undermine the public’s confidence in the democratic process,” they said.
Contact the author
Please read our comment policy before commenting.