Full-text search across every article: title, summary, plain-language explanation, and the complete text — archived stories included. Typos are fine.
415 results for “AI Security”
… Instead of only checking code statically, it starts AI agents that inspect a target, try attack paths, and attempt to prove vulnerabilities with proofs of concept. The reason Strix belongs in this tools special is …
What this is about CAI stands for Cybersecurity AI and is a framework by Alias Robotics for security agents. It targets people who want to structure security testing, CTF tasks, bug bounty analysis, or internal red-team …
What this is about Security company Air Security disclosed a vulnerability it calls Plugin4Shell on September 17, 2026. It affects the plugin mechanisms in Claude Code, OpenAI Codex, GitHub Copilot CLI, and Gemini CLI. …
What this is about Cisco introduced the AI Supply Chain Provenance Explorer on July 30, 2026. According to Cisco, the public tool catalogs almost 900 open AI models and brings together information on provenance, provider …
What this is about Sand Security described a serious flaw in Writer AI on July 7, 2026: an attacker could share a harmless-looking agent preview link. If a logged-in Writer user opened that link, the user's session token …
… feature as a relay, because Bing is allowed by the content security policy. The flow is dangerous because the link can point to a trusted Microsoft domain. According to Varonis, the attacker needs no plugins, no special …
What this is about Reports of AI systems ignoring instructions, bypassing safeguards, or deceiving their users have reached a new high. On August 29, 2026, the Guardian reported, citing the Loss of Control Observatory, …
… is about OpenTaint is an open-source static application-security tool. It tracks how potentially untrusted input flows through source code and whether it reaches dangerous operations without sufficient validation. The …
What this is about Vigolium is an open security tool for web and application scanning. Its core idea is deliberately two-track: a deterministic Native Scan for fast, repeatable checks and an Agentic Scan that combines …
… Report. The report matters because it does not treat AI as a future promise. It looks at concrete attack patterns from real security incidents. The clearest message is that attackers are leaning more heavily on software …
