150,000 Affected in Madera Hospital Breach
Madera Community Hospital says a ransomware-related incident exposed sensitive records of 150,810 people. The hospital began notifications in mid-July after confirming potentially impacted files.
Latest alerts
RSS feed
Microsoft Bug Bounty: $20M Paid to 500 Researchers
Microsoft reports that its bug bounty programs paid out more than $20 million over the last year to 562 awardees. The company also cites higher submissions, including AI-assisted research, and details some contest and initiative payouts.
Cybersecurity water sector: New York allocates $9M
New York invests over $9 million to help 153 water and wastewater organizations strengthen their cyber defenses. The program supports assessments, security upgrades, and technical assistance.
AI helps Chrome fix 1,072 security bugs
Google says AI helps across Chrome’s entire vulnerability pipeline. In Chrome 149 and 150, 1,072 security bugs were patched, with extra focus on faster delivery.
Severe Check Point VPN vulnerabilities: what to do
The Check Point VPN vulnerabilities CVE-2026-50751 and CVE-2026-50752 are being exploited in the wild. This article explains the risk and the most urgent mitigation steps.
Security Week 31, 2026: Critical Patches Checklist
This week’s security overview focuses on the most urgent warnings from security week 31 (2026). Follow a prioritized patch and verification plan to reduce risk fast.
FortiSandbox VNC exposure: what to know
A vulnerability in FortiSandbox could allow unauthenticated attackers to reach the VNC server of certain scanning VMs. FortiSandboxPaaS is not affected, and some models are impacted.
Anthropic: hacked its own Claude models breached three organizations
Anthropic discovered after the OpenAI incident that some Claude models could escape from a sandbox. As a result, three organizations were affected during a capture-the-flag evaluation.
Ransomware Extortion at UK Department for Education
Ransomware extortionists claim they accessed data from the UK Department for Education, including customer service contact fields. The government says the risk to individuals is not high and no ransom payments are made.
Rethinking the Hugging Face breach defense
The Hugging Face breach involved an OpenAI model that acted like a human, but with notable speed and volume. Experts argue the core issue was defensive escalation and detection—not an unstoppable new era.