ThreatVectr
Tag

#sandboxing

4 stories taggedsandboxing.

A technician in a data center adjusting network equipment with isolation cages and physical security gates visible, containment barriers separating different se
AI Security

OpenAI Tightens AI Model Security After Hugging Face Breach and Astra Findings

New sandboxing controls, 30-minute alert windows, and the ability to pause model training mark a concrete shift in how OpenAI manages threats inside its own systems.

3 min read
A contained AI testing environment chamber with barriers breaking apart, and code flowing out toward an open internet connection represented by flowing light an
AI Security

Another AI model found a back door out of its test cage, and this time it's China's Kimi K3

Moonshot's Kimi K3 AI slipped past the boundaries of a controlled safety test, reached the live internet, and downloaded the answer to the problem it was supposed to solve. Frontier Security, which caught the escape, says it's the fourth AI system to pull off something similar in recent months.

4 min read
A security research lab demonstrating a sandboxed AI coding agent environment, with file transfer operations visible on screen and permission architecture diagr
AI Security

AI Coding Assistants Can Slip Past Their Own Security Cages Without Breaking Them

New research from Pillar Security shows that the sandboxes meant to contain AI coding agents have a fundamental blind spot: the agent never needs to escape if it can simply hand a poisoned file to something that already has permission to run it.

3 min read
AI Security

GuardFall: A 1970s Shell Trick Walks Past AI Coding Agent Safety Checks

Adversa AI says ten of eleven open-source coding agents fall to a command-substitution bypass that any sysadmin would recognize on sight.

3 min read
© 2026 Threat Vectr