There’s plenty to be said about the surprising places artificial intelligence turns up—chess tournaments, art contests, your email’s spam folder—but lately, it’s been attempting a (literally) diplomatic career. CNN reports that someone used AI to impersonate Secretary of State Marco Rubio, contacting at least five public figures—three foreign ministers, a US governor, and a member of Congress—all through Signal, the encrypted messaging app. The digital doppelgänger wasn’t just trekking the internet for kicks: the apparent aim was to lure victims into giving up sensitive information or access to accounts.
If that scenario sounds more like a cautionary tech parable than actual news, rest assured, it’s been confirmed by a real-life diplomatic cable. Described by Reuters, the cable specifically warned diplomats to alert their external contacts about fake government officials who, somewhat unnervingly, are now both invisible and synthetic.
A.I., Meet the World Stage
The impersonation effort was both ambitious and oddly old-fashioned: the actor created a Signal account under the name “marco.rubio@state.gov” (nothing sets the mood like a government email sign-off, even a fake one) and left AI-generated voicemails with at least two targeted officials. In at least one case, according to the Reuters-cited cable, the target received a follow-up text message, an invitation to continue the conversation on Signal—presumably, in that familiar mix of bureaucrat-ese and algorithm.
The cable, detailed by Bloomberg, described the purpose behind these overtures: to “manipulate targeted individuals using AI-generated text and voice messages, with the goal of gaining access to information or accounts.” There’s something strangely retro about using modern technology for a classic con—like “phishing” updated for 2025, but with a slightly too-perfect accent and an oddly enthusiastic approach to diplomatic scheduling.
Neither the cable nor State Department officials named a perpetrator, as Reuters underlined. However, a State Department spokesperson confirmed to the press that the agency is “aware of this incident and is currently investigating the matter,” emphasizing ongoing efforts to improve cybersecurity. For security reasons—and, one suspects, for the sake of plausible deniability—no further details were provided on the record.
The Not-So-Remote Control
Adding a layer to this saga, the same diplomatic cable pointed out there’s a second ongoing campaign, this one attributed to a Russia-linked hacker. According to TRT Global and Reuters, that earlier attempt, which started in April, involved phishing messages in which the perpetrator posed as a State Department official (complete with a fake “@state.gov” email address and copied State Bureau logos), targeting the Gmail accounts of think-tank scholars, activists in Eastern Europe, journalists, and former officials. The actor behind this campaign showed “extensive knowledge of the Department’s naming conventions and internal documentation,” the cable noted, with industry partners tying the activity to the Russian Foreign Intelligence Service.
Within this tangled web, the cable stated plainly: “There is no direct cyber threat to the department from this campaign, but information shared with a third party could be exposed if targeted individuals are compromised.” Officials also highlighted that incidents should be promptly reported—either to the FBI’s Internet Crime Complaint Center for outside partners, or to diplomatic security for internal staff.
The New Normal (with a Hint of Absurdity)
While no breaches or secrets swapped hands (as far as anyone’s admitting), the notion of foreign ministers and senior US officials pausing to determine whether they’re speaking to a real person or a well-trained mimic is undeniably strange. Perhaps even more than usual, the diplomatic circuit must now wrestle not just with mistranslation or cultural confusion, but the possibility that the person on the other end is only pretending to exist.
All of which raises a practical question: will updates to “security protocols” soon include pop quizzes about shared coffee break preferences or deeply obscure inside jokes—the only shibboleths left in a sea of algorithmically assembled correspondence? As TRT Global and Reuters both emphasized, there’s a world of difference between an obviously fake Twitter bot account and an AI voice leaving plausible, if unsettling, messages in your voicemail.
There’s a degree of real-world risk here, as underscored by the State Department’s steady warnings. But set against the thoroughly odd landscape of 2025, it’s hard not to marvel at the spectacle—secretaries of state verifying their own voices; politicians second-guessing late-night texts from their supposed colleagues; and, somewhere out there, a bot still practicing its Capitol Hill cadence.
In this era, the line between a botched prank and a genuine state-level threat is thinner than ever, and increasingly, it seems that double-checking who’s on the line is as essential as the message itself. Are we ready for a world in which every diplomatic overture comes with the faint, digitized echo of “Would you like to continue this conversation on Signal?” If nothing else, it’s a reminder that sometimes the real story is stranger than any science fiction—AI Rubio and all.
