Tag: emerging threats
4710 articles
GitHub Copilot Exposes Vulnerability to AI Safety Bypass
Researchers have made a surprising discovery about GitHub Copilot, finding that it can be vulnerable to AI safety bypasses, even when it refuses harmful prompts in isolation. This weakness emerges when the same objective is embedded in a typical multi-turn IDE session.
Hackers Exploit Roundcube Flaw to Target Academic Researchers
A new wave of cyber attacks linked to China is targeting academic researchers in the US and Canada, specifically those in physics, engineering, and national security-related fields, by exploiting a vulnerability in Roundcube webmail servers. The campaign, tracked as 'UNK_MassTraction', has been ongoing since May and has already hit several universities.
AI Coding Agents Trigger Endpoint Security Rules Meant for Attackers
In a surprising twist, over half of the blocked activity detected by Sophos in June 2026 came from developer coding assistants, not hackers, triggering endpoint security rules meant to catch malicious actors. This unexpected behavior highlights the need for a closer look at the intersection of AI-powered coding tools and cybersecurity protocols.
Phishing Campaign Targets Microsoft 365 Users with Voice-Based Entra Passkey Scam
Beware of scammers impersonating Microsoft 365, tricking users into enrolling a fake Entra passkey by mimicking the real enrollment portal and leveraging voice calls to urge action. This sneaky phishing campaign has been targeting multiple sectors since April, putting unsuspecting users at risk.
Anthropic's Claude Code Exposes Security Risk, China Alleges
A Chinese cybersecurity group has raised a red flag about a potential backdoor security risk in Anthropic's Claude Code, warning that certain versions can secretly send sensitive user data to remote servers without consent. This alarming claim puts users' identity and location information at risk.
Accenture Breach Exposes Source Code, Heightens Supply Chain Risk
Accenture's recent data breach, where 35GB of sensitive data including source code was stolen, shines a spotlight on the hidden risks of working with major consulting and services firms. As a trusted partner to businesses and governments worldwide, Accenture's breach heightens concerns about supply chain vulnerabilities.
Multiple Breaches Expose Millions in June
A massive data breach at Madison Square Garden put over 26 million records at risk after a threat actor group, ShinyHunters, made a ransom demand and released the data when it wasn't met. This alarming incident highlights the growing threat of data breaches and the importance of robust cybersecurity measures.
AI Coding Assistants Exposed to HalluSquatting Botnet Attack
Researchers have uncovered a sneaky new attack method called HalluSquatting that targets AI coding assistants, exploiting their tendency to invent names and run code with minimal human oversight. This clever tactic chains together AI behaviors like hallucination and prompt injection to deliver malware efficiently.
RedWing Spyware Targets Android Users via Telegram
Meet RedWing, a sneaky Android spyware that's being rented out as a service on Telegram, targeting unsuspecting users and institutions, with a staggering 82 organizations, mostly Russian financial firms, already in its sights. This malware-as-a-service operation is surprisingly polished, complete with a user-friendly interface, tutorial videos, and even a referral scheme to spread its reach.
Ubiquiti Fixes Flaws in UniFi Ecosystem
Ubiquiti has patched a critical vulnerability in its UniFi ecosystem, specifically a command injection flaw with a perfect 10.0 CVSS score, that could let hackers take control of your device. The update fixes issues across UniFi products, shielding you from potential attacks that could escalate privileges or make unauthorized changes.
AI Coding Agents Expose Unix-Era Security Flaw
A clever trick that exploits a long-standing Unix security flaw, dubbed GhostApproval, can bypass human approvals in AI coding assistants, rendering consent meaningless. By manipulating a harmless-looking project file, attackers can secretly alter sensitive system settings.
China-Linked APT Bolsters Proxy Network with Custom Malware Arsenal
Meet UAT-7810, a China-linked advanced persistent threat that's rapidly expanding its proxy network with custom malware, allowing other attackers to hide their tracks and route traffic through compromised devices. This sophisticated operation, known as LapDogs, has been providing infrastructure for malicious activities for years.
Turkish Defence Firms Penetrate NATO Markets Amid European Rearmament Push
Turkish defence firms are making a major breakthrough into NATO markets, with record exports of over $10 billion in 2025, and sales to Europe and the US nearly quadrupling to $5.6 billion. Turkish companies are now supplying top NATO nations like Poland, Spain, Portugal, and Romania with cutting-edge military hardware.
Türkiye Unveils Tayfun Block-3 Anti-Ship Ballistic Missile Capability
Türkiye has successfully tested its Tayfun Block-3 anti-ship ballistic missile, demonstrating its capability to destroy a moving target at sea with precision and speed. In a live-fire trial, the missile locked onto and struck a seven-metre unmanned vessel at hypersonic speed, showcasing its advanced terminal seeker technology.
SCMBANKER Malware Targets Mexican Banking Users with ClickFix Lures
Mexican banking customers beware: a sneaky new malware campaign, dubbed REF6045, is using fake CAPTCHA pages and social tricks to install a powerful PowerShell toolkit called SCMBANKER on unsuspecting victims' devices. This stealthy attack has been targeting Mexico's financial ecosystem, putting fintech users, payment-processor clients, and cryptocurrency exchange customers at risk.
China Warns of Claude Code Backdoor Risks, Urges Developers to Uninstall
China's National Vulnerability Database has issued a high-priority alert, warning developers to immediately uninstall certain versions of Claude Code due to a potential backdoor risk that could compromise sensitive data. Upgrade to the latest secure version to safeguard your information.
Convicted Felons Launch Offensive Cybersecurity Firm, Lure Researchers with Million-Dollar Payouts
Meet IRIS C2, a bold new cybersecurity firm launched by convicted felons, shaking up the industry with million-dollar payouts to attract top vulnerability researchers and exploit developers. They're offering up to $7 million for zero-day exploits and other cutting-edge capabilities.
AI Models Exacerbate Cybersecurity Skill Gap
The rapid evolution of AI is supercharging cyber threats, making it crucial to update our defenses ASAP - what worked yesterday may not cut it tomorrow. To stay ahead, experts recommend leveraging AI to bolster security and detect vulnerabilities faster than ever before.
AI-Powered Attacks Target Service Desks With Convincing Impersonation Tactics
Beware: AI-powered attacks are now targeting service desks with incredibly convincing impersonation tactics, making it easier for attackers to trick agents into bypassing security controls. A single phone call can be all it takes to spark a devastating data breach, as seen in high-profile incidents at major companies like M&S, MGM Resorts, and Clorox.
US AI Clearinghouse Must Bridge Vulnerability Gap
The US AI cybersecurity clearinghouse has a crucial role to play in bridging the vulnerability gap, but time is of the essence - AI tools are surfacing vulnerabilities at a pace that's outstripping our ability to act on them. With a 30-day deadline now expired, the clearinghouse must swiftly coordinate efforts to scan, discover, and prioritize critical infrastructure vulnerabilities.
Australia Exposed to Uyghur Forced Labour Imports
Australia's imports from China have raised red flags, with a staggering $6.95 billion worth of goods from high-risk sectors linked to Uyghur forced labour flooding the market in 2024. This exposes the country to a redirected risk of inadvertently supporting human rights abuses through its supply chains.
China's PLA Embeds Tactical Air Control Teams
The Chinese People's Liberation Army (PLA) has made significant strides in air-ground integration, with tactical air control teams now seamlessly coordinating with air force aircraft to execute precision strikes. This capability, showcased in recent cross-service training exercises, enables ground teams to guide fighter jets to targets using advanced technology like laser designators and secure datalinks.
DHS Faces Scrutiny Over Unclassified Network Breach
A breach of the Department of Homeland Security's Homeland Security Information Network has sparked concern among lawmakers, who warn that the exposed information, though unclassified, poses a national security risk. The House Homeland Security Committee has requested a briefing from DHS on the intrusion, seeking answers by Friday.
EvilTokens Exposes New Blind Spot in Email Security
A shocking 75.6% of consulting firms were exposed to phishing attacks in 2026, with other industries like financial services, manufacturing, and tech also falling prey to these threats. EvilTokens' ghost phishing campaign uses a sneaky Microsoft Device Code Phishing tactic to trick victims into giving hackers access to their Microsoft 365 accounts.