White House Preparing to Give Federal Agencies Access to Anthropic’s Mythos AI Model
Summary The White House Office of Management and Budget (OMB) is setting up protections to allow major federal agencies access to Anthropic’s closely guarded Mythos AI model, according to a…
CISA Adds Apache ActiveMQ Vulnerability CVE-2026-34197 to Known Exploited Vulnerabilities Catalog
Summary The Cybersecurity and Infrastructure Security Agency (CISA) has added a new vulnerability to its Known Exploited Vulnerabilities (KEV) Catalog: CVE-2026-34197, an Apache ActiveMQ Improper Input Validation vulnerability that is…
Stanford HAI 2026 AI Index: Training Grok 4 Produced 72,816 Tons of CO2 — AI’s Environmental Cost Is Scaling Fast
Summary The Stanford HAI 2026 AI Index Report is out, and the environmental findings are staggering. Training xAI’s Grok 4 model produced an estimated 72,816 tons of CO2 equivalent —…
“Comment and Control” — New Prompt Injection Attack Hijacks Claude Code, Gemini CLI, and GitHub Copilot Agents
Summary Security researchers have disclosed a new prompt injection attack method called “Comment and Control” that can hijack AI coding agents through untrusted GitHub data — PR titles, issue comments,…
Critical Fortinet FortiSandbox Vulnerabilities Patched — CVE-2026-39813 and CVE-2026-39808 (CVSS 9.8)
Summary Fortinet has released emergency patches for two critical vulnerabilities in its FortiSandbox product, both scoring CVSS 9.8. These flaws allow unauthenticated attackers to bypass authentication and execute arbitrary code…
McGraw Hill Data Breach Exposes 13.5 Million Accounts — ShinyHunters Exploits Salesforce Misconfiguration
Summary Educational publishing giant McGraw Hill has confirmed a data breach affecting 13.5 million user accounts, after the ShinyHunters cybercriminal group leaked over 100GB of stolen data. The breach originated…
OpenAI Expands “Trusted Access for Cyber” Program with GPT-5.4-Cyber — A Dedicated AI Model for Defenders
Summary OpenAI has announced a major expansion of its Trusted Access for Cyber (TAC) program, alongside the release of GPT-5.4-Cyber — a specialized variant of GPT-5.4 fine-tuned specifically for cybersecurity…
IBM Announces New Defenses Against “Agentic Attacks” as AI-Powered Cyber Threats Escalate
Summary IBM has announced a new suite of cybersecurity measures specifically designed to help enterprises defend against “agentic attacks” — cyberattacks planned and executed by autonomous AI agents rather than…
Critical nginx-ui Authentication Bypass Under Active Exploitation — CVE-2026-33032 (CVSS 9.8)
Summary A critical authentication bypass vulnerability in nginx-ui, a popular open-source web-based management interface for Nginx servers, is under active exploitation. Tracked as CVE-2026-33032 with a CVSS score of 9.8,…
Pro-Iranian Group Claims Devastating Cyberattack on LA Metro — Alleges 500TB Data Wiped, Rail Systems Accessed
Summary A pro-Iranian threat actor identifying itself as Ababil of Minab has claimed responsibility for a cyberattack against the Los Angeles County Metropolitan Transportation Authority (LACMTA). The group alleges it…