Mail bombing is getting a new look with a wide-scale, ongoing social engineering campaign that might be tied to the Black Basta ransomware gang.
A recent Shai-Hulud attack shows defenders need real-time visibility into production code and affected services to respond to supply chain compromises.
One critical, three high-severity EPICS bugs could disrupt research experiments
The University of New Mexico is working to restore systems following the incident, which has been claimed by the BYOD ransomware group
University hospitals once again accepting patients after initial ambulance diversion, but recovery to continue in phases
Attackers have infected 953 online stores with a scanner-evading JavaScript loader since July 2025
Researchers uncovered an OpenAI API flaw that exposed hidden AI reasoning traces, prompting a fix and a $6000 bug bounty reward
Election messages can ask you to register, donate, take a survey, or buy merchandise. Here’s how to check before you click or pay.
ISC2’s new Cyber-AI Workforce Coalition serves as a “vendor-neutral convener” to map the future of AI security careers, ISC2 CEO Scott Beale told Infosecurity
Both critical PHP flaws land on the agency’s exploited list as infosec pros confirm hackers are getting working shells
Our latest report reveals how AI is reshaping our relationships, blurring what feels authentic, and creating new ways for intimacy to be exploited.
Arctic Wolf warned that the function enables highly effective social engineering attacks, which lead to incidents such as ransomware
Did you hear the one about eMudhra and the power of 4? Time to get connected to certificate trust list updates
PHP and Node.js flaws, hacked sites, fake updates, and rogue certificates can put you at risk even on trusted websites. Here’s how to stay safe.
Interesting research: “ AI Models Can Provably Hide Arbitrary Capabilities ” Abstract: Capability evaluations aim to surface dangerous model behaviors before deployment, but their reliability depends on the assumption that hidden capabilities can be elicited. We show this assumption does not hold under adversarial conditions by constructing backdoor attacks that embed encrypted neural networks…