News

A new agent data injection attack exploits AI agents, causing them to misclick or execute commands from attackers. This highlights vulnerabilities in AI security, emphasizing a need for increased protection against manipulation. Organizations are urged to…

Israeli startup Oak has secured $60 million in seed funding to develop an AI-driven Identity Operating System aimed at consolidating identity governance. Co-founded by Shai Morag and Tal Marom, the platform intends to replace legacy systems by providing…

The article discusses the resurgence of the Daxin malware in Taiwan alongside the Stupig pre-login SYSTEM backdoor, indicating a rise in cyber espionage and malware activities. It emphasizes the need for stronger cybersecurity measures as these threats…

Palo Alto Networks' Unit 42 identified TuxBot v3, a new IoT botnet framework that utilizes AI for code development. Notably, the large language model (LLM) used by the developer left its safety disclaimers in the code, which is present in all sixty-one C…

Splunk and Zoom have issued patches for several vulnerabilities in their products. Splunk addressed three specific high-severity flaws (CVE-2026-20296, CVE-2026-20297, CVE-2026-20298) in versions 10.4.1, 10.2.5, 10.0.8, and 9.4.13, alongside critical issues…

Security updates have been released for various software, including Adobe products, Chrome, Firefox, VMware, and Zoom. Adobe patched critical flaws in ColdFusion, Commerce, and Experience Manager. Google addressed 15 security vulnerabilities in Chrome, while…

The article discusses the effectiveness of artificial intelligence (AI) in identifying software bugs, emphasizing that while AI can detect issues efficiently, human expertise remains crucial for validation and comprehensive understanding. It highlights the…

Samsung recently faced backlash from users of its health app after introducing a consent toggle that threatened to delete users' health data unless they agreed to allow it to be used for AI training. Following significant user pushback and inquiries, Samsung…

The SANS Institute alerts that while AI adoption in cybersecurity surged to 78%, a significant governance gap persists. Interviews with 536 cybersecurity practitioners revealed that 63% faced serious issues in threat detection and response, highlighting…

A critical vulnerability in Shark vacuum cleaners allows attackers to control multiple devices across a region due to an unpatched flaw. This highlights ongoing issues in IoT security, emphasizing the need for stronger protections and awareness around device…

The report highlights a Chinese cyber espionage campaign utilizing Claude Code and DeepSeek to automate attacks targeting government systems and financial firms. Hunt.io researchers discovered this campaign in June 2026 through compromised servers in Hong…

The US government has launched the Gold Eagle initiative to enhance AI-driven vulnerability management, aiming to accelerate the detection and remediation of cyber vulnerabilities. Collaborating with agencies like CISA, the Treasury, and private sector…

Research from Natto Thoughts reveals that China's military procurement system has suspended or banned over a dozen cybersecurity vendors due to contract bidding misconduct. Between 2021 and 2026, there were at least 21 enforcement actions. The penalties…

OpenAI's GPT-Red has been developed to automate prompt injection testing aimed at strengthening the security of the GPT-5.6 model. The initiative focuses on utilizing red teaming techniques and enhancing software security measures to bolster the safeguards…

ESET has warned about vulnerabilities in old Unified Extensible Firmware Interface (UEFI) shim bootloaders signed by Microsoft that could allow attackers to bypass Secure Boot protections. There are 11 UEFI shims identified, mostly from version 0.9 and…

Zoom has addressed a critical vulnerability, tracked as CVE-2026-53412, which allows unauthenticated attackers to take over user accounts on Windows systems with a CVSS score of 9.8. This flaw affects older versions of the Zoom Desktop Client, VDI Client,…

Zoom has released a critical update addressing a vulnerability in its Windows application that could allow attackers to take over user accounts. This patch aims to enhance enterprise security, ensuring safer user interactions and protecting sensitive…

Law enforcement across Spain, Portugal, and the U.S. dismantled a sophisticated €140 million cyber fraud operation involving over 70 individuals, 19 registered companies, and nearly 1,000 financial accounts. The criminal gang executed various cyber scams…

Nightmare Eclipse, a security researcher, has released a new zero-day exploit called LegacyHive that targets a local privilege escalation vulnerability in Windows User Profile Service. This exploit allows attackers to gain access to other users' hives,…

The article discusses recent important updates from cybersecurity companies Trend Micro, ESET, Tenable, and Tanium that address critical vulnerabilities in their products. \n1. **Tenable** fixed a critical path traversal vulnerability allowing potential…

The content provides an overview of the current threat level, which is categorized as 'green'. It features information about the handler on duty, Manuel Humberto Santander Pelaez, and includes links to previous posts and podcasts related to the topic.…

On July 14, 2026, Microsoft discovered a coordinated supply chain attack affecting the @asyncapi npm organization. Malicious versions of five packages were republished, embedding a harmful loader that activated during module imports, bypassing typical npm…

The article discusses the integration of Harden-Runner, a security tool designed to protect third-party GitHub Actions runners, which now includes support for Bitrise macOS runners in version 2.20.0, along with Blacksmith, Depot, Namespace, and Warp. It…

Harden-Runner v2.20.0 introduces egress block mode for macOS and Windows GitHub-hosted runners, enhancing security by preventing secret exfiltration across all supported operating systems. Previous attacks exposed vulnerabilities where compromised workflows…

The article introduces Device Policy, a new feature that allows security teams to enforce a list of approved VS Code extensions across their organization using tools like Intune, Jamf, or the DMG agent. Key points include: 1. **Functionality**: Device Policy…

The article discusses the discovery of 11 vulnerable UEFI shim bootloaders that have been trusted for years, exposing a critical blind spot in Secure Boot systems. These bootloaders, discovered by ESET, could allow attackers to bypass Secure Boot protections…

A recent report by Sophos reveals that identity attacks are now the leading cause of ransomware, surpassing previous reliance on software vulnerabilities. Key findings show that malicious emails (26%) and phishing (24%) constitute the top exploit methods,…

Following a record number of security patches released by Microsoft, a researcher published an exploit called HiveLegacy that allows low-privileged Windows accounts to modify administrator accounts. The exploit targets vulnerabilities in the Windows User…

TELEPUZ is a new modular malware discovered spreading through CLICKFIX-VIDAR attacks. It is lightweight and under active development with a growing presence on VirusTotal. The infection process starts with social engineering tactics that prompt users to…

The US and allied governments have issued warnings regarding Russian state-sponsored Advanced Persistent Threat (APT) groups targeting poorly secured network devices, particularly routers, to compromise critical infrastructure globally. These threats are…

The article discusses the evolution of TuxBot v3, a notable development in IoT botnet capabilities powered by large language models (LLMs). It highlights the implications of this evolution for cybersecurity, particularly in risk containment and the…

Dark Reading has launched a new section called DR Global Europe, targeting cybersecurity professionals in the EU and UK. This initiative aims to provide tailored intelligence and insights specific to Europe's unique cybersecurity landscape, which experiences…