BYOD ransomware plays audio threats on PCs

Staff at two American universities heard the ransom demand spoken aloud from their own computers, in a tactic that turned a normally private extortion note into a public alarm. A new analysis of the BYOD ransomware campaign, published on 20 November, links the audio messages at the University of New Mexico and Vanderbilt University to a wider run of claims against at least nine organisations between 5 October and 9 November.
The researchers behind the analysis, at TrendAI, stress that the disruptions are confirmed while much of what the gang claims is not. They had no public malware sample to examine, and their findings draw on victim statements, public reporting and the group's own claims. That distinction matters, because BYOD's loudest boasts have not been confirmed by the institutions it names.
Ambulances diverted in New Mexico
At the University of New Mexico Health Sciences Center, staff reported hearing ransom messages from their computers during a cyberattack on 26 October. The centre is New Mexico's only Level 1 trauma centre, and it diverted ambulances to nearby hospitals for two hours. More than 30 clinics cancelled appointments and outpatient surgeries, new admissions were temporarily limited, and classes at four health sciences schools were cancelled.
BYOD later claimed it had compromised 36,000 systems at the university, including its Microsoft Entra cloud identity service and two Oracle databases. The university gave a narrower account: its investigation found the attack affected some end-user devices but did not reach core systems or servers. Critical electronic systems were back online the next day, and normal ambulance traffic resumed.
The group also offered free decryption keys to healthcare facilities after the disruption, while saying it would keep the data it claimed to have stolen. That offer does not establish which systems, if any, were encrypted, or what information the attackers obtained.
Vanderbilt computers played the threat aloud
About two weeks later, a recording from Vanderbilt University captured a voice announcing: "You have been compromised by BYOD." On 9 November, the university told students, faculty and staff to shut down university devices and disconnect from campus Wi-Fi while it investigated. Campus Wi-Fi returned that evening, with other services restored in phases.
BYOD claimed it had encrypted Vanderbilt's databases and stolen research data and more than one million Social Security numbers. Vanderbilt confirmed the incident but had not publicly verified the group's involvement, encryption or data theft in the reporting reviewed by TrendAI. A reported ransom note, named README_BYOD_31.txt, threatened further disruption.
The university warned users to watch for unexpected login approval requests and unusual changes to files or permissions in cloud storage. Its advice to anyone hearing an unexpected audio message was direct: switch the computer off and report it from another device.
Nine organisations, and claims that outrun the evidence
TrendAI counted at least nine organisations claimed by BYOD in the five weeks to 9 November. Michigan State University is among them. The university reported a possible credential compromise in early October and required account holders to reset their passwords. BYOD later claimed it had stolen personal information on 1.3 million current and former students. Michigan State said it found no evidence that attackers had compromised its central Active Directory or accessed, altered or stolen other university data.
BYOD also claimed an attack on Texas Health and Human Services on 9 November. That agency had not publicly confirmed an incident when the claim was reported.
The audio messages themselves tell investigators less than they might seem to, the analysis notes. They show that a message played on affected computers, but they do not establish how the group delivered it, whether the same method was used at both sites, or whether attackers controlled a central directory. The public reporting does not identify an exploited software vulnerability.
TrendAI recommends that security teams check identity, endpoint and cloud logs together, then review what a suspected account accessed before its password was reset, including shared files and administrative settings. The campaign's lesson, the researchers say, is to treat the confirmed disruption as real and the gang's figures as claims until a victim confirms them.