ZeroFox Daily Intelligence Brief

ZeroFox Daily Intelligence Brief - September 9, 2026

ZeroFox Intelligence collects, curates, and analyzes information derived from open and proprietary sources. Here is today’s daily roundup to give you and your clients an advantage over the adversary.

Brief Highlights

Ransomware Group Claims 19 Victims in 24 Hours Including LA Metro
ShinyHunters Claims Breach of Florida DMV Driver Records Database
Chinese AI Companies Target U.S. Frontier Models for Knowledge Distillation

Ransomware Group Claims 19 Victims in 24 Hours Including LA Metro

What we know: In the past 24 hours, ZeroFox has observed The Gentlemen ransomware group to be the most active threat actor with at least 19 victims listed on its leak site. U.S. public transit authority LA Metro is among the latest alleged victims. Furthermore, U.S.-based healthcare technology company Veradigm has disclosed a cybersecurity incident, days after The Gentleman claimed to have targeted the entity.

Context:

The Gentlemen has set a nine-day deadline for LA Metro to respond to its demands. So far, it has not published any sample data. LA Metro has not publicly confirmed the claim. Earlier in May, it was reported that LA Metro was reportedly breached by pro-Iranian hacktivist group Ababil of Minab.
Meanwhile, Veradigm confirmed that an unauthorized party used a third-party vendor's credentials to access a Veradigm API serving its customers. No encryption or operational disruption has been reported.The Gentlemen claimed to be in possession of over 3.5 million patient records.
See The Gentlemen’s threat actor profile here.

Analyst note:

The Gentlemen is likely to publish an alleged dataset related to LA Metro if negotiations fail. The targeting of LA Metro by two threat actors within six months likely indicates an unremediated foothold, resold access, or a publicly exposed flaw that drew follow-on intrusion. There is also a roughly even chance that The Gentlemen is likely recycling the data from the March 2026 intrusion.
In the case of Veradigm, the compromised third party vendor is likely to hold integration credentials for other healthcare technology platforms, which the threat group will likely use to reach additional targets. Furthermore, exposed individuals are likely to be targeted in phishing and social engineering attacks for financial theft. Threat actors are also likely to leverage the alleged data for insurance fraud.

ShinyHunters Claims Breach of Florida DMV Driver Records Database

Source: https://cloud.zerofox.com/intelligence_new/advanced_dark_web/110695

What we know: ShinyHunters extortion group has claimed to have breached an online platform for the Florida Department of Motor Vehicles (DMV) database, operated by the Florida Department of Highway Safety and Motor Vehicles (FLHSMV). The department has not confirmed any breach or unauthorized access.

Context: The group released a screenshot of an alleged record from Florida's Driver and Vehicle Information Database (DAVID) containing information of “Jeffrey Epstein,” including a Social Security number (SSN), driver identification number, physical details, and vehicle information. The group claims to have breached the database through a password-reset flaw, allegedly stealing over 200,000 records.

Analyst note: If access to DAVID is confirmed, the breach is very likely to result in downstream risks such as identity theft and financial fraud. The group's use of a high-profile individual's record as proof of access is very likely a deliberate tactic to maximize media attention, as well as an indication that information on other high-profile members is likely in the database.

Chinese AI Companies Target U.S. Frontier Models for Knowledge Distillation

Source: https://www.cisa.gov/news-events/cybersecurity-advisories/aa26-251a

What we know: Chinese AI companies including DeepSeek and Moonshot AI have conducted large-scale knowledge-distillation campaigns against U.S. AI models such as Claude, several of OpenAI's GPT models, and more, CISA has warned. The extracted outputs were reportedly fed into the development and improvement of Chinese AI models.

Context: The China-based AI companies reportedly use a grey market of application programming interfaces (APIs) proxies called “transfer stations” and shared premium accounts to bypass geographic restrictions, evade safeguards while accessing U.S. AI models at scale. Distillation is a legitimate AI development technique, but China’s large-scale, targeted, and unauthorized use to extract capabilities from third-party frontier models represents a misuse of the technique.

Analyst note: This activity is likely to increase the risk of frontier-model capability replication and safeguard evasion, as repeated probing can provide insight into both model behavior and defensive controls, which can be used to develop techniques for bypassing them.

DEEP AND DARK WEB INTELLIGENCE

Telegram user 313 Team: Pro-Palestinian threat group "313 Team" has claimed to have carried out a distributed denial-of-service (DDoS) attack against King Abdulaziz International Airport (AKA Jeddah Airport) in Saudi Arabia. The threat actor stated that the attack disrupted all internal servers and resulted in a complete shutdown of the website. So far, there have been no reports of operational disruption to the airport’s website. It is likely a false claim of impact by 313 Team given its reputation.

DATA BREACHES INTELLIGENCE

Mathspace data breach: Education platform Mathspace has confirmed a data breach affecting 1,079,819 students, parents, guardians, teachers, and employees in Australia and New Zealand. Attackers exploited CVE-2026-72898, a critical SQL injection vulnerability in the company's self-hosted Metabase installation, to gain unauthenticated administrator access and download data from an internal reporting database. The compromised data included names, usernames, email addresses, internal user IDs, countries, time zones, account types, email-verification status, account creation dates, and recent login or activity dates.

VULNERABILITY AND EXPLOIT INTELLIGENCE

Microsoft patch Tuesday September 2026: Microsoft has patched 974 vulnerabilities, including two zero-day flaws actively exploited in the wild. CVE-2026-85880 is a Windows ALPC heap buffer overflow that could enable an attacker running code in a low-privilege AppContainer to escape the sandbox and elevate privileges to System without user interaction. CVE-2026-81963 is an improper link resolution flaw in the Windows Update Stack that could also enable an attacker to elevate privileges to System.

Affected products: The affected products are listed in the advisory.

Dark web chatter: Well-regarded threat actor “TylerDurden” has advertised an alleged unauthenticated remote code execution (RCE) zero-day in Magento 2 for USD 12,500 on dark web forum Exploit. Meanwhile, Moderately credible threat actor “btc1837” has advertised an alleged Hyper-V zero-day flaw that could enable VM escape from a root partition through vmswitch[.]sys and other Hyper-V drivers for USD 100,000–200,000 on Exploit.