12-Year-Old PostgreSQL Flaw Lets Attackers Execute Code and Take Over Database Servers
A critical PostgreSQL vulnerability dubbed PostGREShell could allow low-privileged replication accounts to execute attacker-controlled code, escalate to data...
GBHackers
20 articles
Hackers Use Frontier AI Agents to Breach Enterprise Network in Under 10 Hours
A threat actor used frontier artificial-intelligence models and attack-specific agentic frameworks to breach an enterprise environment, harvest root credenti...
CARS24 Data Breach Exposes 3,100 Customer Records, Leads Allegedly Sold for ₹1,000 Each
Used-car platform CARS24 has alleged that confidential information belonging to approximately 3,100 customers was stolen and supplied to a rival business and...
Hackers Turn HiveMQ and Element Messenger Into Control Channels for Windows Backdoors
The financially motivated threat actor Toy Ghouls has expanded its custom malware arsenal with two Windows backdoors that abuse HiveMQ’s public MQTT infrastr...
OpenAI Agents Collude on Public Wiki to Share Sandbox Bypass and Evasion Techniques
Researchers have discovered a public wiki message board that they claim was used by autonomous AI agents, identifying themselves as OpenAI systems, to exchan...
ShipMonk Data Breach Exposes Personal Data of 67,000 Additional Trezor Customers
Trezor has revealed that a data breach involving its fulfillment provider, ShipMonk, exposed personal and order information of approximately 67,000 additiona...
NodeStealer Spyware Adds Keylogging, Screenshot Capture and Facebook Data Theft
A major upgrade to the Python-based NodeStealer malware, transforming the Facebook-focused infostealer into a broader spyware platform capable of logging key...
Microsoft Teams Adds QR Code Protection to Block Phishing and Fraud
Microsoft is developing a new security feature for Teams messaging that will obscure QR codes sent by external users. This measure aims to help organizations...
MECCHA CHAMELEON Flaw Lets Malicious Custom Maps Achieve Remote Code Execution
A recently patched vulnerability in MECCHA CHAMELEON allowed attacker-controlled Steam Workshop maps to write files to arbitrary locations on Windows systems...
Critical Super Forms WordPress Flaw Actively Exploited to Achieve Remote Code Execution
Threat actors are actively exploiting a critical vulnerability in the Super Forms WordPress plugin, allowing them to upload PHP backdoors and gain remote cod...
Plex Urges Users to Update Media Server as Multiple Security Flaws Are Discovered
Plex has urged users to promptly update their Plex Media Server and Plex Desktop software following the release of fixes for several undisclosed security iss...
Hackers Abuse AI-Era ASCII Smuggling to Hide Phishing Content in Millions of Emails
Threat actors have repurposed an AI prompt-injection technique known as ASCII smuggling to evade email security controls at massive scale, hiding invisible U...
Chinese-Speaking Hackers Use Claude, Qwen and DeepSeek AI Agents to Attack Government Systems
Chinese-speaking threat operators have been observed using Claude, Qwen and DeepSeek-powered AI agents as operational components in a second intrusion campai...
TP-Link Archer AX55 Flaws Enable Remote Code Execution and Admin Password Theft
TP-Link has released security updates for two vulnerabilities found in its Archer AX55 v4 wireless router. These vulnerabilities could allow attackers on the...
Microsoft 365 Direct Send Bypass Lets Attackers Spoof Internal Users Without Credentials
A Microsoft 365 email security-control bypass that lets attackers submit unauthenticated messages posing as internal users by leaving one SMTP field blank. T...
Google Chrome V8 Flaw Actively Exploited in the Wild, Update Released
Google has released an urgent update for Chrome Stable to address CVE-2026-85046, a high-severity type confusion vulnerability in the V8 JavaScript and WebAs...
CrowdStrike Falcon Zero-Day Lets Attackers Escalate Privileges on Windows Systems
A recently released proof-of-concept, named FalconFlank, claims to reveal a local privilege escalation vulnerability in the CrowdStrike Falcon Sensor on Wind...
Contagious Interview Operators Move Beyond Git Hooks With Trojanized Mac Applications
North Korea-linked Contagious Interview operators have expanded their developer-targeting malware delivery operation beyond booby-trapped Git hooks and codin...
Hackers Compromise More Than 14,500 Dahua Security Cameras in Massive Campaign
A large-scale campaign dubbed Operation CameraSwarm compromised at least 14,530 Dahua IP cameras and related surveillance devices in just 35 days. Exposing h...
LLMjacking Attack Abuses Leaked AWS Credentials to Hijack Amazon Bedrock AI Models
Threat actors are increasingly converting stolen cloud credentials into access to costly generative AI services, a technique known as LLMjacking.