Articles and insights dedicated to AI security.
VirusTotal’s recent blog series dissected a new AI‑driven malware framework called OpenClaw. Using machine‑learning models, the tool can automatically craft complex infection techniques such as reverse shells, “semantic worms” that alter code
Read post ->The latest OpenClaw analysis reveals that threat actors are now embedding reverse shells inside AI‑generated scripts, making the payloads appear as legitimate code snippets. These scripts are delivered via compromised supply‑chain packages and
Read post ->VirusTotal researchers identified a new infection chain that abuses the OpenClaw AI platform. Threat actors created dozens of seemingly harmless “skills”—small AI‑driven scripts that claim to perform routine tasks such as file cleanup
Read post ->Cloudflare has launched a new Account Abuse Protection service that sits on its edge network and uses behavioral analytics combined with AI models to continuously evaluate login attempts and account activity. The platform automatically challenges
Read post ->A recent VirusTotal deep‑dive uncovered that the OpenClaw malware family has been upgraded with artificial‑intelligence‑driven modules. The new capabilities include dynamically generated reverse shells that tunnel outbound traffic through encrypted channels, and
Read post ->VirusTotal researchers have uncovered that threat actors are hijacking OpenClaw AI agents and publishing malicious “skills.” These skills instruct compromised endpoints to fetch and execute untrusted code, turning the AI platform into an automated infection
Read post ->Cloudflare has moved its AI‑driven Security for Apps service into general availability. The platform uses large‑language‑model inference at the edge to profile HTTP requests, automatically flagging and blocking malicious traffic such as
Read post ->