- RxRPC Page-Cache Write vulnerability, part of Dirty Frag chain
- Fragnesia Linux kernel privilege escalation vulnerability
- xfrm-ESP Page-Cache Write vulnerability, part of Dirty Frag chain
ThreatNoir Afternoon Brief — May 14
Afternoon Review in IT Security — May 14, 2026
The threat landscape continues to evolve rapidly, with critical vulnerabilities emerging across multiple platforms and threat actors demonstrating aggressive exploitation timelines. Today's security briefing highlights significant risks spanning Linux kernel flaws, Windows zero-days, iPhone security concerns, and artificial intelligence infrastructure vulnerabilities.
New Fragnesia Linux Flaw Lets Attackers Gain Root Privileges
A high-severity kernel privilege escalation vulnerability has emerged in Linux distributions, tracked as CVE-2026-46300 and known as Fragnesia. This flaw allows attackers to execute malicious code with root-level privileges, representing a critical threat to Linux infrastructure worldwide. Source: New Fragnesia Linux flaw lets attackers gain root privileges. Linux distribution maintainers are actively rolling out patches to address the vulnerability, and system administrators should prioritize applying these updates to prevent potential compromise. The vulnerability is tracked alongside CVE-2026-43284 and CVE-2026-43500 in security databases.
Researcher Drops YellowKey, GreenPlasma Windows Zero-Days
A security researcher has publicly disclosed two critical Windows zero-day vulnerabilities that pose significant risks to enterprise and consumer systems. YellowKey functions as a BitLocker bypass requiring physical access to affected devices, while GreenPlasma enables elevation of privileges to System level. Source: Researcher Drops YellowKey, GreenPlasma Windows Zero-Days. The disclosure of these vulnerabilities in the wild creates an immediate window of exposure for Windows users until Microsoft releases and deploys patches. Organizations should implement physical security controls and privilege access management strategies to mitigate the risks posed by these flaws.
Your iPhone Gets Stolen. Then the Hacking Begins
A thriving underground ecosystem has developed to supply criminals with tools and techniques for unlocking stolen iPhones and conducting subsequent attacks against victims' contacts. Source: Your iPhone Gets Stolen. Then the Hacking Begins. Criminals leverage compromised devices to launch phishing campaigns targeting the victim's contacts, with the ultimate goal of accessing financial accounts and sensitive personal information. This threat demonstrates the critical importance of device-level security measures and user awareness regarding post-theft risks.
Hackers Targeted PraisonAI Vulnerability Hours After Disclosure
Exploitation attempts against a PraisonAI authentication bypass vulnerability were observed less than four hours after public disclosure, highlighting the aggressive speed at which threat actors move to weaponize newly revealed flaws. Source: Hackers Targeted PraisonAI Vulnerability Hours After Disclosure. The vulnerability, tracked as CVE-2026-44338, affects artificial intelligence infrastructure and underscores the expanding attack surface in AI-driven environments. Organizations deploying PraisonAI should immediately assess their exposure and apply available mitigations to prevent unauthorized access.
The convergence of these threats across operating systems, infrastructure components, and emerging technologies underscores the importance of maintaining robust patch management programs and threat intelligence monitoring capabilities.
Sources & IOCs
Source articles and extracted indicators (defanged where appropriate).
- YellowKeyWindows zero-day BitLocker bypass exploit requiring physical access
- BlueHammerPrevious Windows Defender security defect exploited by threat actors before patches
- GreenPlasmaWindows zero-day privilege escalation to System level
- iRealmPhishing software that generates Apple service lookalike pages and links to bypass Find My iPhone protections
- Find My iPhone OffPhishing kit used to access Apple iCloud accounts and disable Find My on stolen devices
- Authentication bypass in PraisonAI versions 2.5.6 to 4.6.33 due to disabled Flask API authentication
- CVE-Detector/1.0Scanner conducting reconnaissance of vulnerable PraisonAI endpoints within hours of disclosure