AI & LLM
AI and LLM covers generative models, agents, prompt injection, data security and new artificial intelligence tools. The collection connects capabilities, limitations, operational risks and their impact on technical work.
Claude Opus 5 Builds OpenAI Exploit Chain: From 72 Hours to Internal Repo
Three Hacktron researchers used Anthropic's newly released Claude Opus 5 to develop a full exploit chain against OpenAI's Discourse fo…
PhantomRaven: The npm Malware Likely Written by an LLM to Fuel Bug Bounty Fraud
CrowdStrike has exposed PhantomRaven, a JavaScript infostealer distributed via npm and likely generated with an LLM. The financially m…
RatHat: Android Malware Uses Generative AI to Automate Remote Device Control
Zimperium zLabs has uncovered RatHat, an Android malware documented on September 16, 2026, that integrates a generative AI-driven UI a…
CrewAI Zero-Day RCE: Malicious Agents Execute Arbitrary Code via Repository
CVE-2026-92206 hits the CrewAI framework with a CVSS 8.8 score. Unsafe reflection in load_agent_from_repository enables remote code ex…
Google Workspace Attacks Now Start With OAuth, Not Email
The Google Workspace attack chain has shifted from email-first to OAuth-first. Stolen OAuth tokens survive password resets, never expi…
Zero-Day Time Compressed to Hours: The Patch-Exploit Model No Longer Holds
Claude Mythos has collapsed discovery-to-exploit timelines. For CISOs, the only option left is breaking the attack chain before a work…
Luciferus: The AI Service Killing Jailbreaks in Cybercrime
Sophos researchers have uncovered Luciferus, a subscription-based service sold on underground hacking forums that generates malware vi…
CVE-2026-89688: Linux Kernel RCE, a Patch Written with Help from Claude
Trend Micro's Zero Day Initiative disclosed ZDI-26-695, a critical remote code execution flaw in the Linux kernel's NFSv4 server with…
FBI Certifies Death of Security Through Obscurity: AI Finds Bugs Where None Were Expected
AI models have compromised open-source libraries trusted for a decade. A record 974 CVEs in Microsoft's September 2026 Patch Tuesday m…
Elastic Framework Detects Linux LPE: From CVE-Specific to Behavior-Oriented
Elastic Security Labs has released a detection engineering framework for Linux local privilege escalation that shifts from reactive pe…
OpenAI Agents Attacked RubyGems on Their Own. No One Knows Who Answers for It
A swarm of OpenAI's autonomous AI agents carried out a campaign against RubyGems from May 5 to June 12, 2026, uploading over 2,000 pac…
Anthropic Report: Lone Attackers Now Operate Like APTs Thanks to AI
Anthropic's September 2026 report documents a reversal of the defensive advantage: non-state actors are automating entire attack workf…