UK PM Spoofed in Wiles Impersonation; Rubio AI Lures Hit 3 Ministers

For security teams, this is what a top-tier whaling attack looks like at the state level. British PM Andy Burnham reportedly messaged an impersonator of White House chief of staff Susie Wiles before becoming suspicious, despite the White House later saying no Wiles devices were compromised. With the FBI and State Department already warning about AI-driven impersonation of senior officials, CISOs and threat-intel teams should treat this as proof that identity deception—not device hacking—remains the most dangerous vector against executives and governments.

On August 17, 2026, multiple outlets including ABC7, KSL, KSAT and The Morning Call reported that British Prime Minister Andy Burnham had exchanged messages with someone impersonating White House Chief of Staff Susie Wiles. Politico first reported the incident citing four unnamed officials. Burnham, who has been in office less than a month, reportedly believed he was messaging Wiles before becoming suspicious and cutting off communications. A Burnham spokesperson declined to comment, citing policy not to discuss national security matters, and the White House said the episode had nothing to do with Wiles' devices being hacked.

“

British PM Andy Burnham reportedly messaged an impersonator of White House chief of staff Susie Wiles before becoming suspicious, despite the White House later saying no Wiles devices were compromised.

The incident is not an isolated anomaly. It follows a broader pattern of high-level impersonation campaigns documented by US government agencies. Last year, the US government investigated a series of messages that elected officials, business executives and other prominent figures received from someone posing as Wiles. Shortly afterward, the State Department warned US diplomats of attempts to impersonate Secretary of State Marco Rubio and possibly other officials using artificial intelligence. An impostor posing as Rubio attempted to reach out to at least three foreign ministers, a US senator, and a governor. The FBI also warned of malicious actors misusing AI to impersonate senior US government officials. These prior events provide critical context: the Burnham case appears to be a continuation or escalation of a known VIP impersonation campaign, not a one-off breach.

From a security standpoint, this is best understood as a social engineering attack rather than a technical compromise. The White House explicitly stated that Wiles' devices were not hacked, meaning the attacker likely relied on spoofed messaging accounts, fraudulent identities on messaging platforms, or compromised third-party channels. The target was not data but trust. For any security team, the takeaways are significant: identity verification of high-level correspondents, out-of-band confirmation for sensitive discussions, and continuous briefing of executives and political principals about impersonation tactics. The fact that a sitting head of government could be drawn into a false exchange for a period before suspicion arose underscores how difficult even experienced officials find it to challenge a trusted persona.

The geopolitics amplify the severity. Burnham took office less than a month ago and has been actively trying to forge good ties with the White House after Trump's relationship with his predecessor, Keir Starmer, soured. That transition window offers a perfect opportunity for social engineering: new principals establishing communication norms, eager to cultivate key relationships, and with less established personal channels. An attacker could exploit the urgency and desire for access. The reported exchange is an attempt to collect intelligence, shape policy, or insert disinformation at the highest level of government-to-government communication. Because the British PM's office has declined to comment beyond a national security policy, the full content of the messages, the platform used, and whether any actionable intelligence was lost remain unknown.