17+ hour, 13+ min ago  (464+ words) Prompt injection and data theft risks in agentic LLM browsers are turning routine web bugs into full-session compromise paths, enabling attackers to hijack AI agents and exfiltrate sensitive data at scale silently. As agentic browsers like Perplexity Comet, Open AI…...

1+ day, 13+ hour ago  (233+ words) Open AI has unveiled GPT-5. 4-Cyber, a specialized variant of its newest AI model designed exclusively for defensive cybersecurity. The launch accompanies a major expansion of the Trusted Access for Cyber (TAC) program. This controlled initiative provides verified security professionals…...

2+ day, 13+ hour ago  (13+ words) SAP Patch Day Fixes Critical SQL Injection, Do S, and Code Injection Flaws'cyberpress. org...

6+ day, 8+ hour ago  (487+ words) A critical vulnerability in Git Hub Copilot Chat has revealed a new and dangerous way attackers can silently steal sensitive data. The flaw, tracked as CVE-2025-59145 with a CVSS score of 9. 6, allowed hackers to exfiltrate secrets such as API keys…...

6+ day, 13+ hour ago  (408+ words) A critical vulnerability in a popular Android communication library recently exposed millions of digital asset holders to severe data theft risks. Security researchers discovered a redirect flaw in Engage SDK, a widely used third-party tool for managing push notifications. This…...

6+ day, 14+ hour ago  (468+ words) A newly uncovered jailbreak technique dubbed "sockpuppeting" is raising fresh concerns across the AI security landscape after researchers demonstrated that a single line of code can bypass safety guardrails in 11 leading large language models (LLMs), including Chat GPT, Claude, and…...

1+ week, 15+ hour ago  (291+ words) Git Lab has released critical security updates addressing multiple vulnerabilities affecting both its Community Edition (CE) and Enterprise Edition (EE), including flaws that could enable denial-of-service (Do S) attacks and code injection. The latest patch versions 18. 10. 3, 18. 9. 5, and 18. 8. 9 fix a total of…...

1+ week, 1+ day ago  (382+ words) Anthropic has announced "Claude Mythos Preview," a new AI model that dramatically raises the stakes in software security by autonomously finding and exploiting zero-day vulnerabilities in widely used systems. The company is positioning the model as both a breakthrough defensive…...

1+ week, 3+ day ago  (468+ words) Google is rolling out a notable update to its Chrome browser, introducing native lazy loading support for audio and video elements, a move that not only boosts web performance but also strengthens privacy and security controls across modern websites. The…...

1+ week, 3+ day ago  (476+ words) The Apache Software Foundation has released emergency security updates to address two high-severity vulnerabilities in Apache Traffic Server (ATS), a widely deployed, high-performance web proxy and caching solution used across enterprise environments. Disclosed on April 2, 2026, the flaws could allow remote,…...