(TNND) — Someone reportedly used artificial intelligence-powered software to impersonate Secretary of State Marco Rubio in a series of voice and text messages to foreign ministers, a governor and a member of Congress, The Washington Post reported.

The newspaper cited an unnamed senior U.S. official and a State Department cable.

The Post said that officials haven’t identified the culprit but believe they were trying to gain access to information or accounts.

The Rubio impostor created a Signal account with a fake email address for the secretary of state.

“The actor left voicemails on Signal for at least two targeted individuals and in one instance, sent a text message inviting the individual to communicate on Signal,” said the cable, according to The Post’s story.

A senior State Department official told The National News Desk via an emailed statement that they are investigating the matter.

“The Department takes seriously its responsibility to safeguard its information and continuously takes steps to improve the department’s cybersecurity posture to prevent future incidents,” the official said in the statement. “For security reasons, and due to our ongoing investigation, we are not in a position to offer further details at this time.”

AI expert Anton Dahbura said the Rubio impostor was using technology that has emerged over the last several years.

AI is now publicly available that can capture a person’s voice from a video or audio clip and then generate an impersonation.

And an even more dangerous form of that technology is emerging that could allow a scammer to impersonate someone else’s voice in real time.

“It's an indicator of things to come,” said Dahbura, the co-director of the Johns Hopkins Institute for Assured Autonomy. "It's what we call reality hijacking."

RELATED STORY: Report takes wide-ranging look at AI and its 'absolutely astonishing' pace of development

The FBI warned in May of bad actors using text messages and AI-generated voice messages to impersonate senior government officials.

The FBI said the scammers would send their victims a malicious link under the guise of transitioning the conversation to a separate messaging platform.

Dahbura said the Rubio impersonation involved hacking and social engineering efforts.

“It takes a bit of effort to fool high-ranking government officials in the U.S. or elsewhere. But it's really a warning for everyone,” Dahbura said. “It's not that difficult to get someone's phone number and call them and say that you're from their bank or the FBI or the sheriff's office or anything and perpetrate scams. So, it's taking scams to a whole new level.”

The platforms that allow people, often overseas, to generate an AI-powered voice impersonation work in just seconds.

And Dahbura said the safeguards on those sites aren’t enough.

He also said these incidents show the need for federal AI regulations.

RELATED STORY: Experts, public differ on AI's future; agree on need for regulation, responsibility

A bad actor can make a fake recording with just a bit of audio they grab from social media, for example.

It’s even easier to impersonate a public figure like Rubio, with his voice on hours of video clips posted across the internet.

“It's a pretty tall order to be able to counteract that,” said Dahbura, mentioning audio watermarks as one potential tool.

MORE: Millions of web page visits examined to see how AI touches our online lives

Dahbura said AI, like other technology, can be used for both good and bad.

But it might become increasingly difficult to keep a lid on malevolent uses.

"Seeing is believing. Hearing is believing. That's our last bastion of defense for authenticating who we're interacting with, and that's pretty much gone,” Dahbura said.