What Happened
On January 21, 2024, two days before the New Hampshire Democratic primary, thousands of registered Democratic voters received robocalls featuring a convincing AI-generated imitation of President Biden’s voice. The fake Biden told voters to “save your vote for the November election” and not to participate in the primary, stating that voting in the primary would only help Republicans. The calls were designed to suppress Democratic voter turnout.
The voice clone was created using ElevenLabs’ voice synthesis technology. Investigation traced the calls to political consultant Steve Kramer, who was working for rival Democratic candidate Dean Phillips. Kramer later claimed he orchestrated the scheme to draw attention to AI deepfake threats, though prosecutors rejected this defense.
Why It Matters
This was the first major instance of AI voice cloning being weaponized in a U.S. election. It demonstrated that the tools to create convincing political deepfakes were accessible, affordable (reportedly costing around $1), and effective enough to potentially influence elections. The incident occurred at the start of a presidential election year, raising alarms about the vulnerability of democratic processes to AI-enabled disinformation. It prompted the fastest regulatory response to an AI incident in U.S. history, with the FCC issuing a declaratory ruling within weeks.
Lessons Learned
AI voice cloning technology has reached a quality threshold where it can convincingly impersonate political figures. The cost barrier to creating deepfakes is effectively zero. Existing telecommunications regulations were inadequate to address AI-generated voice fraud. Regulatory frameworks must proactively address AI misuse rather than responding after harm occurs. Voice authentication and deepfake detection tools need to be integrated into telecommunications infrastructure.
Current Status
The FCC unanimously declared AI-generated voice robocalls illegal under the Telephone Consumer Protection Act in February 2024. Steve Kramer was fined $6 million by the FCC and faced criminal charges in New Hampshire. ElevenLabs updated its terms of service and detection tools. The incident accelerated bipartisan support for AI regulation in election contexts. Multiple states passed laws specifically addressing AI deepfakes in political communications.