A malicious AI “skill” created as part of a controlled security experiment has exposed critical weaknesses in modern AI agent ecosystems, successfully bypassing security scanners and compromising more than 26,000 agents across individual and enterprise environments. According to researcher Niv…
Tag: Cyber Security News
Critical Cisco Unified CM and SME Flaw Enables Remote Attacker to Launch SSRF Attacks
Cisco has warned customers about a critical server-side request forgery (SSRF) flaw in Cisco Unified Communications Manager (Unified CM) and Unified CM Session Management Edition (Unified CM SME) that allows remote, unauthenticated attackers to write files on the underlying OS…
How Attackers Exploit Privileged Access and How to Lock Them Out
Every major breach you read about has a quiet middle chapter that rarely makes the headline. The headline is the ransom note or the leaked customer database. The middle chapter the part that actually decided the outcome is almost always the same: an attacker found a…
FortiBleed Attack Hit 430,000+ FortiGate Firewalls, Stealing 110M+ Credentials
A large-scale, ongoing credential-harvesting campaign dubbed “FortiBleed” has silently compromised more than 430,000 FortiGate firewalls globally, siphoning over 110 million credentials directly from live network traffic since at least February 2026. The campaign came to light after security researcher Volodymyr…
GTA 6 Scam Websites Use AI-Generated Images and Fake Download Buttons to Lure Gamers
A fresh wave of scam websites is targeting gamers worldwide, using the massive hype around Grand Theft Auto VI to trick people into handing over their money. These fake pages promise something millions of players desperately want: early access to…
CISA Warns of Ubiquiti UniFi OS Vulnerability Actively Exploited in Attacks
The U.S. Cybersecurity and Infrastructure Security Agency (CISA) has added multiple Ubiquiti UniFi OS vulnerabilities to its Known Exploited Vulnerabilities (KEV) catalog, warning that at least one of the flaws is now being actively exploited in the wild. Federal civilian…
In-Browser Data Inspection Lets Analysts Track Phishing Attack Flow Inside Browser Sessions
Phishing attacks have grown far more complex in recent years. Attackers no longer rely on simple static pages to steal credentials. Instead, they build layered redirect chains, execute dynamic scripts, and load content in stages, making it much harder for…
Hackers Use GoogleErrorReport Scheduled Task for Persistence in Dropping Elephant Campaign
A well-known threat actor called Dropping Elephant has returned with a refined and more dangerous campaign, using a China-themed lure document to drop a reworked remote access trojan (RAT) onto victim machines. The attack is designed to stay hidden, avoid…
Claude Down – A Major Outage Affects Most of the Models
Anthropic experienced a service disruption on Tuesday that produced elevated error rates across multiple Claude models, according to the company’s official status page. By mid-afternoon UTC the company said a fix had been deployed and that it was monitoring systems…
AWS Warns Outbound Traffic Blind Spots Can Enable Cloud Data Exfiltration
Most organizations spend a lot of time locking the front door of their cloud environments. Firewalls, access controls, and web application filters get the bulk of attention because that is where visible threats tend to show up. But what leaves…
Bajaj Auto Confirms Systems Affected by Ransomware Attack
India’s leading two-wheeler manufacturer, Bajaj Auto, disclosed on Tuesday that it fell victim to a ransomware attack that compromised systems at both the parent company and its wholly owned technology subsidiary, Bajaj Auto Technology Ltd (BATL). The cybersecurity incident was…
Your SOC Has Too Many IOCs: How to Cut Feed Noise, Prioritize What Matters, and Improve Response
Most SOCs measure threat intelligence the same way they measure storage: bigger is better. A feed that delivers two million indicators a month looks more impressive on a vendor scorecard than one that delivers two hundred thousand. Dashboards proudly display…
Anthropic Launches Claude Tag – AI Teammate Now Lives Inside Slack
Anthropic has unveiled Claude Tag, a new agentic AI feature that integrates directly into Slack, allowing teams to tag @Claude as a collaborative team member to delegate tasks, automate workflows, and build shared organizational context. The feature is available today…
Nearly Half of Apps Across LG and Samsung TV’S are Selling Your IP Address
New research found that 2,058 of 6,038 apps across the LG webOS and Samsung Tizen ecosystems included residential proxy SDKs, effectively turning smart TVs into exit nodes for third-party internet traffic. On screen, these apps look like harmless fish tanks,…
Five-Eye Agencies Call for “Whole-of-Organization and Whole-of-Society Response” to Stop Cyber Threats
The Five Eyes cyber security agencies have issued a joint warning urging governments, businesses, and critical infrastructure operators to adopt a “whole-of-organization and whole-of-society response” to address rapidly evolving cyber threats driven by artificial intelligence (AI). In a statement released…
DifyTap Flaws Allow Attackers to Wiretap AI Data Across Tenants – 1M+ Apps Impacted
Multiple critical vulnerabilities in Dify could expose sensitive AI data across tenants and potentially impact more than one million applications. Dify, which powers AI workflows, chatbots, and retrieval-augmented generation (RAG) pipelines, is heavily adopted across enterprises including Volvo, Maersk, Panasonic,…
LastPass Customer Data Exposed in Klue Supply Chain Attack
LastPass has disclosed a supply chain security incident involving its third-party vendor, Klue, that resulted in unauthorized access to customer data within its Salesforce environment. The company confirmed that the breach did not affect its core infrastructure or password vaults.…
8-Year-Old Samsung KNOX Vulnerability Exposes Galaxy Devices to Kernel Attacks
A critical use-after-free (UAF) vulnerability in Samsung’s proprietary KNOX security subsystem, which has been hidden for over eight years, has been discovered by security research firm LucidBit, potentially exposing hundreds of millions of Galaxy devices to kernel-level memory corruption and…
15 Best Linux Network Monitoring Tools in 2026
The “Linux network monitoring” concept describes keeping monitors on and evaluating a network’s performance, capacity, and overall health. Specialist tools and software capture, measure, and analyze data on network traffic, bandwidth utilization, latency, and connected devices. Administrators can monitor everything…
Scattered Spider Hackers Who Breached London Transport Network Plead Guilty
Two members of the Scattered Spider cybercriminal group have pleaded guilty to a cyberattack on Transport for London (TfL) that caused major service disruptions and resulted in an estimated £29 million in losses. Thalha Jubair, 20, from East London, and…