AI Deepfake Scam Impersonates Marco Rubio in High-Level Cyberattack

In a chilling development highlighting the growing threat of AI abuse, a malicious actor used artificial intelligence voice-cloning to impersonate Secretary of State Marco Rubio and reach out to high-ranking government officials including three foreign ministers, a U.S. governor, and a member of Congress.

According to a July 3 State Department cable, the impersonator leveraged Signal, a secure messaging app commonly used by government officials, to leave voicemails and send texts while posing as Rubio. He even spoofed other State Department personnel via email in a coordinated phishing campaign.

Authorities have not yet identified the perpetrator, but early assessments suggest the goal was to gain access to sensitive information or secured accounts.

This isn't an isolated incident. In May, someone successfully breached the phone of White House Chief of Staff Susie Wiles, using her identity to call multiple top-tier officials. These escalating tactics have prompted FBI warnings to government agencies about AI-powered phishing and impersonation scams targeting officials and their networks.

“If you receive a message claiming to be from a senior U.S. official, do not assume it is authentic,” the FBI advised.

Experts say it only takes 15 to 20 seconds of audio to convincingly clone a voice with today’s AI tools. Once cloned, hackers can generate voicemails, real-time calls, or even video deepfakes with near-perfect accuracy.

Key takeaways:

• The attacker used a fake email to set up a Signal account and sent messages to targeted officials.

• Voicemails using Rubio’s voice were left for at least two individuals.

• The FBI had already warned about a surge in AI-powered phishing schemes earlier this year.

• A recent study found over 50% of phishing emails are now generated by AI.

This is no longer the stuff of sci-fi. AI is now being used in state-level intelligence operations, private scams, and espionage campaigns many of them targeting the United States and its allies. Canadian and Ukrainian agencies have already issued similar warnings after intelligence agencies began mimicking officials to manipulate civilians and recruit informants.

Cybersecurity analyst Hany Farid, a professor at UC Berkeley, summed it up “These attacks don’t require advanced hacking skills just a little creativity and recklessness from the people being targeted.”

The State Department says it is conducting a full investigation and implementing new security measures, but the deeper concern remains: Are our leaders prepared for the AI threat that’s already here?

If you believe in protecting our nation’s leaders and institutions from emerging threats, share this article and subscribe to our newsletter for more updates on AI-driven dangers to national security.