Both Chrome and Firefox have released urgent security updates patching over seventy vulnerabilities combined, including critical memory safety bugs that could allow hackers to remotely execute malicious code. Chrome's update fixes thirty-three security flaws, with seven rated critical severity, most of them use-after-free vulnerabilities that could enable attackers to break out of the browser's sandbox. Firefox patched forty vulnerabilities including thirteen high-severity bugs, and neither company has reported these flaws being actively exploited yet, so users should update their browsers immediately.

Microsoft is working on a patch for a vulnerability in Windows Defender dubbed "RoguePlanet" that allows attackers to exploit a race condition and gain System-level privileges on Windows 10 and 11 machines. The zero-day flaw was publicly disclosed last week by security researcher Nightmare Eclipse, who has recently released several exploits targeting Microsoft products out of frustration with the company's vulnerability disclosure process. The vulnerability, tracked as CVE-2026-50656 with a severity score of 7.8, works regardless of whether Defender's real-time protection is enabled or disabled.

Oracle has released its second monthly Critical Security Patch Update, delivering 245 patches across its product portfolio, including Fusion Middleware, MySQL, and PeopleSoft. The update addresses roughly 120 critical vulnerabilities, with 100 flaws exploitable remotely without authentication, though Oracle hasn't confirmed any zero-day exploits despite recent reports of cybercriminals targeting a PeopleSoft vulnerability that affected at least 100 organizations. The company continues urging customers to apply patches promptly, warning that many successful attacks occur because organizations fail to install available security updates.

The DragonForce ransomware group has deployed a sophisticated new backdoor that disguises its malicious traffic as legitimate Microsoft Teams communications, marking the first known malware to abuse Microsoft's TURN relay infrastructure in this way. The custom malware, called Backdoor.Turn, obtains anonymous Teams tokens and uses legitimate Microsoft relay servers to communicate with attacker command-and-control servers, making it nearly impossible for security tools to distinguish malicious activity from normal Teams traffic. The backdoor was used in a December attack on a US services firm where hackers gained kernel-level access, terminated security processes, and maintained persistent access even after deploying ransomware for data encryption and theft.

Rockwell Automation has released patches for multiple vulnerabilities affecting its industrial control systems, including Logix and CompactLogix controllers, along with its FactoryTalk automation software. The flaws range from critical authentication bypasses and denial-of-service issues to a vulnerability that could allow attackers to take over device accounts by changing web interface passwords. While the company recently confirmed that an older vulnerability has been exploited in the wild, these newly patched security holes have not yet been targeted by threat actors.

Tenet Security has emerged from stealth mode with six million dollars in seed funding to address a critical gap in AI security: detecting and stopping dangerous behavior from AI agents in real time. Founded by former Cisco AI Defense researchers, the company uses its own AI agents to monitor and predict potentially harmful actions before they occur, whether from runaway automation or "agentjacking" attacks where adversaries manipulate agents to act within their authorized permissions. The platform has already proven valuable in early deployments, including blocking attacks at a billion-dollar legal enterprise and catching a runaway agent that racked up tens of thousands of dollars in unnecessary costs over a single weekend.

1Password has acquired Israel-based Apono in a deal reportedly valued between 250 and 300 million dollars, significantly expanding its identity security capabilities. Apono specializes in just-in-time access governance, providing temporary, narrowly scoped permissions for humans, machines, and AI agents that automatically expire after tasks are completed, eliminating the need for persistent privileges. The acquisition allows 1Password to evolve beyond password management into a comprehensive access layer, completing its unified access platform that now covers credentials, SaaS applications, device trust, and privileged access governance.

In a significant supply chain attack, 144 Mastra npm packages were compromised after attackers hijacked a contributor's account. The breach highlights ongoing security vulnerabilities in the JavaScript package ecosystem, where compromised developer credentials can lead to widespread distribution of malicious code. This incident underscores the critical need for stronger authentication measures and monitoring of package repositories that millions of developers rely on daily.

Security researchers have uncovered two dangerous campaigns targeting developers and AI users: malicious JetBrains plugins designed to steal API keys for AI services, and Chrome extensions that capture conversations with AI chatbots. These attacks specifically target valuable credentials and sensitive data from users working with artificial intelligence tools and development environments.

Organizations are increasingly seeking to move beyond basic security visibility to more confident threat prioritization through adversarial exposure validation. This approach helps security teams identify and focus on the vulnerabilities that pose the greatest actual risk to their systems, rather than simply cataloging potential issues. The methodology bridges the gap between detecting security problems and understanding which ones truly matter for an organization's specific environment and threat landscape.