The State Division is warning U.S. diplomats of makes an attempt to impersonate Secretary of State Marco Rubio and probably different officers utilizing know-how pushed by synthetic intelligence, in accordance with two senior officers and a cable despatched final week to all embassies and consulates.
The warning got here after the division found that an impostor posing as Rubio had tried to achieve out to at the least three international ministers, a U.S. senator and a governor, in accordance with the July 3 cable, which was first reported by The Washington Submit.
The recipients of the rip-off messages, which had been despatched by textual content, Sign and voice mail, weren’t recognized within the cable, a replica of which was shared with The Related Press.
“The State Division is conscious of this incident and is at the moment monitoring and addressing the matter,” division spokeswoman Tammy Bruce advised reporters. “The division takes severely its accountability to safeguard its info and repeatedly take steps to enhance the division’s cybersecurity posture to stop future incidents.”
She declined to remark additional resulting from “safety causes” and the continued investigation.
It’s the newest occasion of a high-level Trump administration determine focused by an impersonator, with an identical incident revealed in Could involving President Donald Trump’s chief of workers, Susie Wiles. The misuse of AI to deceive individuals is prone to develop because the know-how improves and turns into extra broadly accessible, and the FBI warned this previous spring about “malicious actors” impersonating senior U.S. authorities officers in a textual content and voice messaging marketing campaign.
The hoaxes involving Rubio had been unsuccessful and “not very refined,” one of many officers mentioned. Nonetheless, the second official mentioned the division deemed it “prudent” to advise all staff and international governments, notably as efforts by international actors to compromise info safety improve.
Study Extra About Securing AI at SecurityWeek’s AI Danger Summit – August 19-20, 2025 on the Ritz-Carlton, Half Moon BayAdvertisement. Scroll to proceed studying.
The officers weren’t approved to debate the matter publicly and spoke on situation of anonymity
“There isn’t a direct cyber menace to the division from this marketing campaign, however info shared with a 3rd get together might be uncovered if focused people are compromised,” the cable mentioned.
The FBI has warned in a public service announcement a couple of “malicious” marketing campaign counting on textual content messages and AI-generated voice messages that purport to return from a senior U.S. official and that goal to dupe different authorities officers in addition to the sufferer’s associates and contacts.
This isn’t the primary time that Rubio has been impersonated in a deepfake. This spring, somebody created a bogus video of him saying he wished to chop off Ukraine’s entry to Elon Musk’s Starlink web service. Ukraine’s authorities later rebutted the false declare.
A number of potential options have been put ahead in recent times to the rising misuse of AI for deception, together with felony penalties and improved media literacy. Issues about deepfakes have additionally led to a flood of recent apps and AI programs designed to identify phonies that would simply idiot a human.
The tech corporations engaged on these programs at the moment are in competitors in opposition to those that would use AI to deceive, in accordance with Siwei Lyu, a professor and pc scientist on the College at Buffalo. He mentioned he’s seen a rise within the variety of deepfakes portraying celebrities, politicians and enterprise leaders because the know-how improves.
Only a few years in the past, fakes contained easy-to-spot flaws — inhuman voices or errors like further fingers — however now the AI is so good, it’s a lot tougher for a human to identify, giving deepfake makers a bonus.
“The extent of realism and high quality is growing,” Lyu mentioned. “It’s an arms race, and proper now the mills are getting the higher hand.”
The Rubio hoax comes after textual content messages and telephone calls went to elected officers, enterprise executives and different distinguished figures from somebody who appeared to have gained entry to the contacts in Wiles’ private cellphone, The Wall Road Journal reported in Could.
A few of those that acquired calls heard a voice that appeared like Wiles, which can have been generated by AI, in accordance with the newspaper. The messages and calls weren’t coming from Wiles’ quantity, the report mentioned. The federal government was investigating.
Study Extra About Securing AI at SecurityWeek’s AI Danger Summit – August 19-20, 2025 on the Ritz-Carlton, Half Moon Bay
Associated: The Wild Wild West of Agentic AI – An Assault Floor CISOs Can’t Afford to Ignore
Associated: Past GenAI: Why Agentic AI Was the Actual Dialog at RSA 2025
Associated: How Hackers Manipulate Agentic AI With Immediate Engineering
Associated: How Agentic AI will likely be Weaponized for Social Engineering Assaults
Associated: Mitigating AI Threats: Bridging the Hole Between AI and Legacy Safety
