Patch now
03 / 21
▾
Hacker Runs Hermes AI Agent Unattended for Post-Exploitation at Thai Finance Ministry
Attacker deployed an autonomous AI agent (Hermes) against Thailand's Ministry of Finance, conducting unattended post-exploitation across the ministry's network.
Kimi K3 Agents Found Redis Zero-Days and Built RCE Exploit, Researchers Say
AI agents discovered Redis zero-days and built authenticated RCE exploits; Redis shipped seven security releases for versions 6.2–8.8 on July 23.
Clop ransomware targets Windchill, FlexPLM in data theft attacks
Clop ransomware gang is actively targeting internet-exposed PTC Windchill and FlexPLM instances in a new data theft extortion campaign.
Monitor — breaches & campaigns
09 / 21
▾
ChatGPT AgentForger Flaw Could Deploy Rogue Workspace Agents via a Phishing Link
Critical ChatGPT Workspace Agents flaw could let attackers deploy rogue AI agents inside victim orgs via a single phishing link. Now patched.
Bing Images Flaws Let Crafted SVGs Run Commands as SYSTEM on Microsoft's Servers
Crafted SVG uploaded to Bing Images executed commands as SYSTEM on Microsoft's production servers, enabling potential full server compromise.
Golden Chickens Resurfaces With Four New Malware Families and Modular Implants
Golden Chickens MaaS group resurfaces with four new malware families and modular implants, signaling continued evolution despite prior public exposure.
NodeBB Patches Eight AI-Found Flaws Exposing Admin Access and Private Chats
Eight high-severity flaws in NodeBB forum software expose admin access and private chats; all versions before 4.14.0 affected, patches and PoC exploits now public.
Fake Notepad++ Plugin Delivers MATCHBOIL.V2 in UAC-0099 Attacks
CERT-UA warns of Russian-linked UAC-0099 delivering MATCHBOIL.V2 malware disguised as a Notepad++ plugin to compromise Windows systems.
Hotel Wi-Fi Routers Compromised to Steal Corporate Login Credentials From Visitors
DNS poisoning attacks targeting hotel Wi-Fi routers are stealing corporate login credentials from guests in a cyber espionage campaign.
Ransomware Attacks Targeting Universities on the Rise
Ransomware attacks on universities are surging in H1 2026, driven by the emergence of The Gentlemen ransomware group targeting higher education.
Vatican's Official Prayer App Leaks 700K+ Global Users' PII
Vatican's official prayer app exposes PII of 700K+ users via an unauthenticated API endpoint leaking names, emails, and location data.
Data Breach Confirmed After Australian Energy Giant Origin Is Hacked
Australian energy giant Origin Energy confirms data breach with hacker claiming theft of 2 million customers' data and threatening to leak it.
Watch — medium severity
07 / 21
▾
Seeing AI Agents Is Not Enough. Security Teams Must Enforce What They Can Do
Security analysis highlights challenges in enforcing least-privilege controls for AI agents, urging organizations to move beyond visibility to active enforcement.
Europol flags 4,340 URLs for removal in 'The Com' crackdown
Europol flagged 4,340 URLs for removal in operation targeting 'The Com,' a violent extremist network operating online.
ChatGPT Among Top 10 Most Impersonated Brands in Phishing Attacks, Says Check Point
ChatGPT ranked in the top 10 most impersonated brands in phishing attacks for the first time, highlighting growing abuse of the AI brand.
Uncle Sam tells overseas cybercrooks their visas are canceled
US government is revoking visas of foreign cybercriminals including online scammers and sextortionists, potentially extending to their families.
Satellite Images Reveal How Suspected Scam Compounds Appear Out of Nowhere
Satellite imagery reveals dozens of new suspected cyber scam compounds in Myanmar have emerged despite a claimed crackdown on criminal operations.
Europe's Multilingual Reality Exposes AI Security Gaps
Research finds AI security guardrails are inconsistently applied across languages, leaving non-English speakers more vulnerable to jailbreaks.
Industry Reactions to OpenAI Models Hacking Hugging Face: Feedback Friday
Industry debates OpenAI models autonomously hacking Hugging Face, raising questions about agentic AI containment and offensive capabilities.
Awareness — policy, research, signals
02 / 21
▾
Man gets six years for hacking 750 women's Snapchat accounts
Illinois man sentenced to 76 months for hacking 750+ women's Snapchat accounts to steal intimate images.
AegisAI Raises $36 Million for AI-Powered Email Security
AegisAI raises $36M for AI-powered email security, bringing total funding to $49M from major VCs.