AI reads 36 sources every 2 hours and ranks each story by importance from 1 to 5 - so you see what actually matters.
Attackers chained three zero-day vulnerabilities in Ivanti's Connect Secure VPN appliance to achieve root-level access on targeted devices. Only two of the three flaws have patches available so far, leaving one exploitable path unaddressed. Because Connect Secure is widely deployed for enterprise remote access, successful exploitation can give intruders a foothold deep inside corporate networks.
SAP disclosed a maximum-severity (CVSS 10) vulnerability in its Cloud SDK for Python that could let attackers steal OAuth2 credentials and gain access to connected cloud services. The flaw affects applications built with the SDK, potentially exposing sensitive business data and enterprise systems. SAP has released fixes, and organizations using the library should update immediately.
GitHub has hardened its npm publishing process after researchers found a flaw that left more than 1,500 packages open to hijacking by attackers. The issue could have let malicious actors take over maintainer accounts or publish rogue versions of widely used libraries. The change is meant to protect the software supply chain that millions of projects rely on.
Xandr has been fined a record €8.6 million for relaying Grindr user data that could reveal sexual orientation to advertising partners. The penalty highlights growing regulatory scrutiny of ad-tech companies handling sensitive personal data under GDPR. It matters because such data can expose vulnerable users and set a precedent for privacy enforcement.
The Hacker News is hosting a webinar on preventing AI-driven privilege escalation using runtime identity controls. It explains why traditional access controls can fail against AI-powered attacks and what security teams can do instead. The session is aimed at security professionals evaluating identity and access management strategies.
Security researchers have found vulnerabilities in LG smart TVs that let an attacker on the same local network gain root-level control of the device without any authentication. Because the flaw requires local network access rather than remote internet exposure, the risk is highest in shared or compromised networks such as hotels, dorms, or offices. The findings highlight the persistent security weaknesses in internet-connected consumer devices.
Bitdefender has published a walkthrough of its AI Visibility and Control offering, which is designed to let organizations see and govern how AI tools are used across their environments. The piece highlights the growing need for enterprises to monitor employee and system use of AI services as adoption spreads.
Attackers remotely took over Danfoss refrigeration controllers at French supermarkets, driving fridge temperatures up to 50°C (122°F). The incident highlights the vulnerability of internet-connected industrial cooling systems used in food retail, where tampering can spoil stock and create safety and health risks.
Two critical vulnerabilities in the MarkLogic database could let users with delegated manager permissions escalate to persistent administrative control. The flaws matter because they may undermine access controls and allow long-term privileged compromise in affected deployments.