AI Deepfake Impersonation Campaign Targeting Senior U.S. Government Officials (2023)

Incident Summary

The FBI issued public service announcements in May 2025 and December 2025 warning of an ongoing malicious messaging campaign in which threat actors impersonated senior U.S. government officials — including White House staff, Cabinet-level officials, members of Congress, and state governors — using AI-generated voice messages and text messages (vishing and smishing).^[1][2]

The campaign, active since at least 2023, targets officials’ personal contacts, family members, and professional acquaintances to extract sensitive information, gain access to protected systems, and conduct financial fraud.^[1][4] The FBI notes that attackers are deploying AI voice-cloning tools that closely replicate the cadence, tone, and accent of real officials.^[2]

Key Facts

• Campaign duration: Active since at least January 2023 through present (as of December 2025)
• Targets: Senior U.S. government officials, their personal contacts, family members, and professional acquaintances
• Officials impersonated: White House staff, Cabinet-level officials, members of Congress, state governors
• Methods: AI-generated voice messages (vishing) and text messages (smishing)
• Objectives: Extracting sensitive information, gaining access to protected systems, financial fraud
• Law enforcement response: Two FBI/IC3 public service announcements issued

Threat Patterns Involved

Primary: Deepfake Identity Hijacking — AI voice-cloning technology used to impersonate senior government officials with sufficient fidelity to deceive their personal contacts and professional associates.

Secondary: AI-Morphed Malware — The campaign uses AI-generated content as a vector for social engineering attacks aimed at compromising government networks and extracting sensitive information.

Significance

1. Government officials as direct targets. This is the first incident in the database where senior government officials and their associates are the direct targets and victims of AI-enabled impersonation, rather than merely appearing as regulators or policymakers.
2. Persistence and scale. The campaign has been active for over two years across multiple branches and levels of government, suggesting a well-resourced and adaptive threat actor.
3. Trust exploitation. The attackers specifically target the personal and professional networks of officials, exploiting established trust relationships that are difficult to protect through technical measures alone.
4. Primary-source documentation. The incident is documented through two FBI/IC3 public service announcements, providing authoritative primary-source evidence of the threat.