Two security researchers bypassed login controls for two automotive brands using default administrator credentials.

OpenAI’s GPT-6 Cyber aims to help approved security teams find flaws. Mozilla has released fixes for many vulnerabilities researchers identified.

OpenAI Introduces GPT-6 Cyber With Enhanced Safety Controls

Remote software vendor N-able suspended its hosted N-central platform globally after an independent researcher released…

Google unveiled Gemini 4 Carbon for cybersecurity and extended coding, initially limiting access to selected cyber defenders.

Google introduced Gemini 4 Carbon for cybersecurity and engineering work, first offering it to selected cyber defenders.

Prince told Reuters that Cloudflare is aware of "a bunch of additional severe vulnerabilities that are present in other pieces of critical software," though he did not name the affected programs. He...

Bank of America has hired 1,000 outside contractors to fix software vulnerabilities uncovered by AI models like Anthropic's Mythos, CNBC reports.

Google has unveiled Gemini 4 Argon, its most advanced AI model yet, but only a handful of cybersecurity partners get access first. The model can autonomously find and patch software vulnerabilities that previous frontier models missed, and it is already saving Google 300 terabytes of memory internally. The broader public rollout has no confirmed date, and the launch comes after months of internal delays and leadership shake-ups. https://qz.com/google-gemini-4-argon-ai-model-cybersecurity-100126

A security researcher discovered nine vulnerabilities impacting ATM encryption and authentication software, but the issues extend beyond ATMs to affect critical systems across industries. Matt Burch's research on CryptoPro Secure Disk, used in ATMs and other embedded devices, revealed flaws that could bypass integrity checks and grant full access to encrypted devices. CryptWare patched the vulnerabilities in two phases, but the broader challenge lies in the software supply chain. Developers must release patches, companies must implement tailored fixes, and customers must install updates, which is difficult for deployed systems. The case underscores the risks of relying on 'security through obscurity' and the need for transparency as AI tools make vulnerability detection easier. While some industries have improved patch adoption, niche security products remain under scrutiny. The incident highlights the complexity of addressing software flaws in a globally distributed supply chain, where delayed updates can leave systems vulnerable to exploitation.

Critical software vulnerabilities exploited, impacting Microsoft and open-source tools; urgent patching needed. #Cybersecurity #SoftwareVulnerabilities #PatchManagement

Cybersecurity researchers have disclosed multiple security vulnerabilities in four popular Microsoft Visual Studio Code (VS Code) extensions that, if successfully exploited, could allow threat actors to steal local files and execute code remotely. The extensions, which have been collectively installed more than 125 million times, are Live Server, Code Runner, Markdown Preview Enhanced, and

Meta has warned users about the critical CVE-2025-27363 vulnerability in FreeType and has emphasized the need for swift patching to prevent exploitation.

Learn about Continuous Vulnerability Management and how DeployHub Pro uses SBOM and Deployment data to detect new vulnerabilities across all assets in your infrastructure from code to cloud.

Learn about Continuous Vulnerability Management and how DeployHub Pro uses SBOM and Deployment data to detect new vulnerabilities across all assets in your infrastructure from code to cloud.

Learn about Continuous Vulnerability Management and how DeployHub Pro uses SBOM and Deployment data to detect new vulnerabilities across all assets in your infrastructure from code to cloud.

Learn about Continuous Vulnerability Management and how DeployHub Pro uses SBOM and Deployment data to detect new vulnerabilities across all assets in your infrastructure from code to cloud.

China's Cyberspace Administration, Ministry of Public Security, and Ministry of Industry and Information Technology introduced.

Tool proliferation has created a data volume challenge, making it harder to secure the software development and deployment lifecycle. Fortunately, AI now offers a solution.

Researchers discover large supply-chain attack targeting Chrome extension developersDozens have been compromised, resulting in possibly millions of victim usersResearchers urge users to patch or uninstall...

Researchers from A Security uncovered a critical vulnerability in Zoom's screen-sharing protocol that could allow attackers to hijack devices during video calls. The flaw, discovered using publicly available AI models, required fewer than 20 prompts to exploit, enabling silent attacks without user interaction. Zoom issued patches for all supported operating systems (Windows, macOS, Linux, iOS, Android) after the vulnerability was disclosed. The issue stemmed from real-time annotation features in screen sharing, which AI systems identified as a common source of overlooked vulnerabilities in proprietary software. A Security cofounders warned that the democratization of AI hacking tools lowers the barrier to entry for malicious actors, emphasizing Zoom's trust-based usage model as an attractive target. Attackers could potentially take control of enterprise networks by joining a call and exploiting the flaw, raising concerns about the security of widely used video conferencing platforms. The incident highlights the growing role of AI in both discovering and exploiting software vulnerabilities, intensifying the 'cat and mouse game' between security researchers and hackers.

As attackers ramp up their AI exploit development, the search for software vulnerabilities is changing rapidly.

#Anthropic’s #Mythos model, unveiled in April, has significantly improved the ability to detect #softwarevulnerabilities. While AI tools are not yet capable of fixing these bugs, they are proving invaluable in identifying them, potentially shifting the balance of power in cybersecurity towards defenders. https://techcrunch.com/2026/05/07/how-anthropics-mythos-has-rewritten-firefoxs-approach-to-cybersecurity/?eicker.news #tech #media #news