This week’s Department of Know is hosted by Rich Stroffolino, with guests Brett Conlon, CISO, American Century Investments, and Michael Bickford, former CISO, New York State Gaming Commission.
Missed the live show? Check it out on YouTube.
The Department of Know is live every Friday at 4:00 p.m. ET. Join us each week by registering for the open discussion at CISOSeries.com.
In this week’s cybersecurity news…
Insurers move to cap LLMjacking cyber payouts
The Financial Times reports insurers including QBE Insurance and Beazley are moving to cap payouts for AI-related cyber incidents, introducing sublimits that significantly restrict coverage for risks like “LLMjacking,” where attackers exploit enterprise AI systems to avoid usage fees. Brokers and legal experts warn the changes could narrow protection across a broader range of emerging AI threats, even as insurers argue they are clarifying coverage rather than reducing it. (Financial Times)
Unauthorized Mythos access, Firebox bugs fixed by Mythos
Speaking of Anthropic, couple different Mythos stories today. Bloomberg reports a small group of unauthorized users claimed in a private Discord that they were able to access Anthropic’s Mythos model, which has been limited to 40 organizations because of its reported ability to find or exploit security vulnerabilities. One member of the group works for a third-party contractor for Anthropic. They combined that access and other information they gathered, like guessing the URL based on how Anthropic formats them. Anthropic is investigating the report and says it has no evidence that the access went beyond a third-party vendor’s environment.
Meanwhile, Mozilla said it used Mythos to look for bugs in Firefox 150, and it found 271. All of the bugs could also have been found by a human, Mythos simply found them faster.
London hospitals continue to suffer from 2024 ransomware attack
A ransomware attack that occurred in June 2024 by the Qilin ransomware group continues to reverberate. Internal documents show at least one NHS trust is still working without fully restored systems and managing large backlogs of delayed test results, restricted blood supplies, the theft and publication of sensitive patient data and delayed treatment of highly time sensitive conditions like cancer. As an example, clinicians at the South London and Maudsley NHS Foundation Trust were “warned not to rely on the timely return of blood results.” Critical results are being communicated by phone, while full reports are being delivered as paper or PDFs and manually uploaded into patient records. A recent study by King’s College London described ransomware as “the most significant current cyber threat to the NHS and warned that a single major technology failure could have serious consequences for patient safety.”
Apple fixes iOS flaw exploited by the FBI
Apple has released an urgent iOS update to fix a security flaw that was reportedly used by the FBI to recover deleted messages. The issue wasn’t in apps like Signal itself, but in the iPhone’s notification system, which stored message previews even after messages were deleted or the app was removed. Investigators were able to access these remnants through the device’s internal database. Apple patched the vulnerability in its latest updates to prevent this kind of data recovery. The case highlights how system-level data can persist beyond user expectations, raising ongoing concerns about privacy, encryption, and how “deleted” data is actually handled on modern devices.
(ZDNet)
Huge thanks to our sponsor, ThreatLocker
of Zero Trust Network Access and Zero Trust Cloud Access, access isn’t based on
credentials alone, it requires the right user, the right device, and the right conditions.
Because as we’ve seen in recent large-scale CRM breaches, stolen credentials and
misconfigurations can expose massive amounts of data. With ThreatLocker, nothing is
exposed, and access is limited to exactly what’s needed. Learn more and start your free
trial today at ThreatLocker.com/CISO.
Vercel confirms breach, stolen data for sale
Vercel disclosed an incident involving unauthorized access to internal systems after an employee’s Google Workspace account was compromised at AI platform Context.ai, then pivoted into Vercel environments. The attacker accessed some customer environment variables that weren’t marked as sensitive and not encrypted at rest, enabling further escalation, though core services and open-source projects like Next.js weren’t impacted. The company is investigating with incident response experts, advising customers to rotate secrets, and review configuration. A threat actor claiming to be ShinyHunters is attempting to sell alleged stolen data.
‘Contagious Interview’ scams self-propagate
According to research from Trend Micro, North Korean threat actors are evolving the “Contagious Interview” scam into a self-propagating supply chain attack, using fake job offers to trick developers into running compromised code that spreads malware through repositories. The campaign is attributed to the group Void Dokkaebi and uses malicious VS Code tasks and hidden repository files to deploy RATs, steal credentials, and infect downstream projects when code is shared. This can rapidly cascade across open-source and enterprise environments, with more than 750 infected repositories now identified.
A novel new concern for CISOs: AI generated ghost breaches
An interesting story appeared this week in Cyberscoop. AI-generated “ghost breaches” are emerging as a new cyber risk: false but convincing breach stories that trigger real-world crisis responses. The article highlights cases where entirely fictional incidents were reported as real, old resolved breaches resurfaced as “new,” and AI-generated quotes were falsely attributed to experts. These narratives can waste security resources, damage reputations, influence regulators and investors, and even help attackers make phishing or impersonation campaigns more believable. The authors argue that CISOs must expand beyond traditional threat intelligence to include “narrative intelligence,” monitor how their organizations are portrayed externally, coordinate closely with communications teams, and conduct regular AI audits to detect false stories early.