Anthropic and EPFL show AI mind viruses can spread through persistent agent prompt files, while a one-paragraph warning cuts spread to near zero.

A suspected China-nexus actor exploits CVE-2026-59310, compromising an estimated 361 IPs in 47 countries and gaining root ...

Cavern uses DNS A records to switch between direct HTTPS and Google Apps Script for C2, expanding an Iranian-linked toolkit ...

Attackers exploit MLflow CVE-2026-64849 via SSRF to steal cloud credentials, while FUXA CVE-2026-25895 draws malicious scanning.

City Forum uses one server to pull records from over-permissive Salesforce and ServiceNow guest portals across industries for over a year.

Autonomous identity governance continuously reassesses access as usage, roles, and risk signals change, replacing stale quarterly reviews.

An unpatched Unisoc modem flaw lets code with a VoLTE foothold modify Android kernel memory on T606, T612, and T7250 chipsets ...

Evooo1Bot exploits known flaws to infect Linux edge devices, then adds SOCKS5 proxying, SSH brute force, credential theft, ...

A weekly look at exploited flaws, exposed systems, supply-chain attacks, browser abuse, malware campaigns, and the security risks that mattered most.

A recently patched security flaw in Apple macOS has come under active exploitation in the wild to deploy a cryptocurrency ...

MCP servers can leak enterprise secrets through plaintext config files or prompt injection. Learn the main risks and how to ...

IAM compliance works only when organizations verify access controls across applications and infrastructure, not just document ...