Today on CISO Series…
Security You Should Know:”Keeping Humans in Charge of AI-Era Defense with ESET”
In todays <a href="https://bitcomme.com/government-it-leaders-to-tackle-ai-cybersecurity-modernization-at-red-hat-symposium-oct-28/” title=”Government IT Leaders to Tackle AI, Cybersecurity, Modernization at Red Hat Symposium Oct. 28″>cybersecurity news…
Microsoft Teams to get AI-powered, context-aware phishing detection
Microsofthas announced that Teams will start using AI-powered phishing detection, with a rollout starting in early November of this year. General worldwide availability is expected by mid-December. This AI-powered phishing detection technology “will use conversation context without retaining message content to identify phishing and suspicious behavior in higher-risk messaging scenarios, with a focus on external first-contact chats.”
Researchers find a third PHP2Shell bug
Following up on the PHP2Shell story, researchers fromThe PHP Foundationhave found a third PHP vulnerability that “could allow attackers to take over websites or remotely run malware on vulnerable systems. But this one won’t receive an emergency fix.” The team has reserved a CVE identifier (CVE-2026-115560) for this third RCE flaw and will release a full advisory and fixed software with the next regular PHP release. This third flaw “requires a very unlikely combination of configuration and specific PHP application code, which we do not expect to occur in practical use of PHP,” they said, and as such doesn’t pose the same massive risk as last week’s other two.
Quest Diagnostics confirms data breach including exposed patient records
The U.S. based clinical laboratory company that provides diagnostic testing and services for a wide range of medical conditions, has disclosed that patient records were exposed following unauthorized access to its Quanum LSM system in mid-October. The LSM system, Lab Services Manager, is used by Quest clients such as physicians, hospital personnel, and medical office staff to place test orders, look up results, and order supplies. Currently, the Quanum LSM system has been halted, and the company is recommending that customers use paper requisitions or Quanum eLabs to place orders and receive results. The company also believes that some patient records were exposed. No one has yet claimed responsibility for this breach.
Senators say targeted ads are a security risk. Ad groups disagree
The Citizen Labat the University of Toronto, working with the cybersecurity-tech firmConfiant Inc, published a report last week on AdFection, a technique in which targeted ads were delivering zero-click mercenary spyware. The report urges ad vendors to “significantly level up their cybersecurity and threat detection capacity to effectively cut off malicious advertisers across the supply chain.”
Democratic Sens. Ron Wyden of Oregon and Cory Booker of New Jersey, who in 2023, introduced legislation to ban advertising networks and facilitators from using personal data to target, legislation that never advanced past introduction, both say that The Citizen Lab report illustrated the need for industry reform.GoogleandIAB Europeresponded by stating that they already work to discover and eliminate malicious ads, but that “the potential misuse of existing technologies and functionalities that are not specific to digital advertising but inherent to web technologies.”
Big thanks to our sponsor,Guardsquare
Bipartisan bill urges federal guidance for distillation attacks
A new bipartisan House bill would “charge the federal government with developing formal guidance for defending AI models against distillation attacks.” Named the AI Cyber Defense Act, and shared exclusively withCyberScoopahead of release, is sponsored by Reps. Pat Harrigan, R-N.C., and George Whitesides, D-Calif. It would urge “National Institute of Standards and Technology (NIST)to develop voluntary industry standards and best practices for preventing distillation attacks from foreign entities from Russia, China, Iran and North Korea.” In an interview, Harrigan told CyberScoop it is “incredibly expensive” for the U.S. AI industry to maintain its six-to-12 month lead over other countries, and distillation attacks give other parties an incentive to copy other models while keeping their own spending low.”
CISA adds Collabora Online and Arista VeloCloud flaws to KEV
The two vulnerabilities impactCollaboraOnline Built-in CODE Server and Arista VeloCloud Orchestrator (VCO). Both have been reported as having active exploitation in the wild. The Collabora vulnerability (CVE-2026-110447) has a CVSS score of 9.2 and is an operating system command injection vulnerability in Collabora Online when used within the NextCloud platform. TheArista NetworksVeloCloud Orchestrator (CVE-2026-114625) has a CVSS score of 9.9 and is an SQL injection vulnerability.
Autonomous attack agent in Skynet mode hijacks other AI models
Researchers from theSymantecThreat Hunter Team discovered an AI agent designed to “hijack exposed AI models to run autonomous attacks, with instructions allowing it to carry out up to 100,000 steps without human approval.” Captured in a honeypot, the agent’s instructions showed how it was “designed to scan the entire internet for endpoints that accept OpenAI-style requests without authentication – or that accept a stolen key – and then to exploit several common web frameworks, along with carrying out dozens of illicit actions.” Within this activity is a malicious loop called “Skynet mode” that allowed the AI agent to run unattended for up to 100,000 steps.
Data breach takes a bite out of crime stoppers organization
A Toronto area crime prevention program used for people to send anonymous tips had to halt operations late Wednesday, following a news report that stated more than 1.5 million tips from Crime Stoppers chapters across the country had been exposed in a data breach. Although Crime Stoppers urges people to not provide their names when submitting information, reporters at Canada’sCBCbroadcaster “were able to reach more than a dozen tipsters by reviewing the hacked data made available to them. TheCanadian Crime Stoppers Associationand individual chapters, blamed a third-party vendor for a lack of information on the breach, but the Toronto chapter decided to come forward with a timeline “showing what it knew and when.”
