Threat intel5 min read
BYOD extortion group plays spoken ransom demands on hospital and campus computers
A new group called BYOD hit UNM Health in October and Vanderbilt in November 2026, playing a synthetic voice on infected computers. The outages were real. The data theft it claims is still unconfirmed.
4Tify Team

A ransomware and extortion group calling itself BYOD has found a loud way to announce itself. During attacks on the University of New Mexico Health Sciences Center in October and on Vanderbilt University in November 2026, computers inside both institutions started speaking. A synthetic voice told staff their data had been taken and named the group. Neither institution has confirmed the data theft BYOD claims. The disruption, however, was real.
Despite the name, the group has nothing to do with "bring your own device" policies. It surfaced in September 2026 and has since claimed universities and health care organisations on its leak site.
What happened at UNM
On Monday 26 October, UNM Health declared a cybersecurity event. Staff were told to shut down and unplug their computers. Ambulances were diverted away from UNM Hospital in Albuquerque, the state's only Level 1 trauma centre, and appointments at more than 30 clinics were cancelled for the day. By 27 October the hospitals were fully open again and ambulance traffic was back to normal.
While the incident was unfolding, a computer-generated voice played on workstations in the hospital. It claimed the attackers held student and employee records, Social Security numbers, documents and email, and told administrators to get in touch through an encrypted messenger. BYOD later claimed on its leak site that it had compromised 36,000 machines, two Oracle databases and the university's entire Microsoft Entra tenant.
UNM's account is much narrower. The Health Sciences Center said the event "affected some end-user devices but did not penetrate core systems or servers". The university said it had made no concessions to the attackers, and it asked anyone the group contacts to report it to the FBI's Internet Crime Complaint Center.
What happened at Vanderbilt
On the morning of Monday 9 November, Vanderbilt said it was responding to an active incident that could affect university devices, systems and cloud services. Students, faculty and staff were told to shut down university devices and disconnect from campus Wi-Fi. The university listed warning signs: unexpected pop-ups, unfamiliar applications, missing or modified files, odd changes to cloud file permissions and authentication requests nobody expected.
Video from campus shows a university computer announcing that the system had been compromised by BYOD. A ransom note named README_BYOD_31.txt was photographed on a screen. It pointed to the group's leak site and gave the Session messenger as the contact channel. BYOD claims it locked databases, took over a domain controller and stole more than a million Social Security numbers.
Wi-Fi came back that evening, and from 10 November people could use their devices again while services were restored in stages. Vanderbilt has not confirmed BYOD's involvement, any encryption or any data theft.
The Michigan State thread
Michigan State University offers a clue about how these intrusions may begin. On 1 October, MSU made everyone with a NetID reset their password to one of at least 15 characters after what it called a potential credential compromise. It said current evidence did not point to a compromise of its central Active Directory, and that it had no evidence other data was accessed. BYOD put MSU on its leak site on 1 November and claimed data on 1.3 million current and former students. That claim is still unverified.
Why the voice matters
A spoken message proves something ran on the affected machines. It doesn't show how the attackers got in, how far they reached or whether anything was stolen. What it does is pressure: patients, students and staff hear the demand and record it before the institution knows what happened. No malware sample, hashes or network indicators have been published, so defenders can't hunt for BYOD by signature yet.
What to do now
- Treat a credential compromise as a possible start of an extortion attack. After any forced reset, review what the affected accounts did beforehand: sign-ins, MFA approvals nobody expected, new admin roles, bulk downloads and changes to sharing settings.
- Watch for unexplained audio playback, new scheduled tasks and remote tools on endpoints. Correlate identity, endpoint and cloud logs instead of reading each one alone.
- Keep offline backups and test restoring them. Hospitals should rehearse downtime procedures that work without the network.
- Give staff, students and patients one trusted place to report messages from anyone claiming to hold their data, and tell them not to reply.
- Work through the CISA #StopRansomware guide for prevention and response checklists.
Groups like BYOD post their claims, and often stolen credentials, on dark web leak sites and forums. Our team's Deep/Dark Web monitoring watches those places for an organisation's domains and accounts, so a leaked login or a leak-site listing reaches you before a voice on a workstation does.