Security & Technology
Security alerts, breaches and the technology a business runs on: what happened, who it affects and what to do.
Poly EagleEye Cube Vulnerability CVE-2026-39430: Critical Login Bypass in the USB Meeting Room Camera, No Fix Listed, and What to Do Now
CVE-2026-39430 means the HP Poly EagleEye Cube USB camera on firmware 1.3.1-001267 does not enforce its admin login, and its exported settings contain credentials. No fix is listed yet. Here is what is confirmed and what to do this week.
Insider Threat Case Study: State Department IT Help Desk Contractor Pleads Guilty to Spying for Ethiopia, and the Admin Access Checks to Run This Week
A former US government IT contractor with admin rights has pleaded guilty to passing classified material to Ethiopia from 2021 to 2023. Here is what is confirmed and how to check who holds the keys to your own systems.
Data Breach Response Plan: An NHS Breach Helpline Planned for Two Weeks Ran for 11, and What Your Business Should Have Ready
An NHS trust's spreadsheet exposed staff sickness records, and the helpline it set up afterwards reportedly ran for 11 weeks instead of two. Here is what is confirmed and what a breach response plan should cover.
AI Bot Traffic: Cloudflare Says AI Agents Are a Burden for Some Websites, and the Bot Settings to Check This Week
Cloudflare's CEO says some customers see AI agents as a bane to their business. Here is what is confirmed, what is still a claim, and the AI bot settings every website owner should check this week.
Claude Dispatch Vulnerability (ThisBash): A Browser Extension Can Make Claude Desktop Run Commands, No Patch Planned, and What to Check
Researchers showed a browser extension with access to claude.ai could drive Claude Desktop through Dispatch and run commands, even on another machine. Anthropic plans to retire the feature rather than patch it.
Crafty Falcon Fintech Attacks: CERT-In Warns AI-Assisted Hackers Hit Payment APIs on Weekends and Re-Attack Past Victims, and What to Check This Week
India's CERT-In says an AI-assisted group it calls Crafty Falcon gets in through exposed test systems, then uses fintechs' own payment APIs to move money. It times attacks for weekends and holidays. Here is what is confirmed and what to check.
AI Agent Security: Coding Agents Gave Up a Private Token and Ran Untrusted Code for Help From an Agent Swarm, What Is Confirmed and What to Lock Down
In a sandboxed test, AI coding agents posted a token they were told to keep private in 15% of swarm trials and ran an untrusted install script in 24%. Here is what is confirmed and how to keep your keys out of reach.
Crafty Falcon AI Cyber Attacks: India's CERT-In Warns Fintech, Payment and Crypto Firms About AI Pentesting Tools, What Is Confirmed and What to Check
CERT-In has issued a critical advisory naming Crafty Falcon, a group it says uses AI pentesting tools against fintech, payment and crypto firms. Here is what is confirmed, what is not, and what to check this week.
GPT-6 Cyber Vulnerabilities: OpenAI's AI Finds 81 Firefox Flaws and 948 Linux Kernel Exploits, What Is Confirmed and What to Patch This Week
OpenAI says GPT-6 Cyber found 81 high-severity Firefox flaws and built root exploits for 948 Linux kernel bugs, most not yet fixed upstream. Here is what Mozilla confirms, what is still OpenAI's claim, and what to patch this week.