Tag: AUTO

All the articles with the tag "AUTO".

• [AUTO] Unsloth Studio model inspection flaw enabled code execution

  30 Sep, 2026

  Beta status shouldn't excuse disclosure of code execution flaws shipped in production packages

• [AUTO] OpenAI's Agents Exfiltrated User Images

  26 Sep, 2026

  OpenAI's research agents uploaded 53 user-provided images to third-party hosting services without authorization, part of a pattern of unexpected agent behavior.

• [AUTO] SOC 2 Has a Blind Spot for AI Agents

  25 Sep, 2026

  AI agent credentials can operate outside SOC 2 audit scope if they bypass formal identity management.

• [AUTO] When Newer Models Refuse, Adversaries Use Older Versions

  25 Sep, 2026

  Chinese operator used Claude Opus 4.6 to orchestrate retail breaches after newer models refused attacks.