» Tag
security
518 postsLinearAutoDestroy: A New Semantic Behavior Type
LinearAutoDestroy is a new semantic behavior type that ensures automatic destruction of values.
Transform Google DeepMind's AI Control Roadmap into Five Security Tests
Google DeepMind's AI Control Roadmap proposes five tests to ensure system safety.
Codex Tightens Sandbox Enforcement for Memory Consolidation
Codex enhances security by preserving parent sandbox enforcement during memory consolidation. This change is crucial for untrusted tasks.
CommitBrief — AI code reviews, right in your terminal
A provider-agnostic, local-first CLI that reviews your staged changes, a historic range, or a whole GitHub pull request. Zero telemetry, no server. Free and open source.
commitbrief.comAI agents can modify themselves without human intervention
AI agents possess the ability to modify themselves autonomously, raising new security concerns for enterprises.
OpenClaw Completes Security Audit
OpenClaw has completed a security audit with Trail of Bits, addressing all identified vulnerabilities.
Adobe Commerce/Magento CVE-2026-75650: Unauthenticated RCE via GraphQL
CVE-2026-75650 in Adobe Commerce/Magento presents a serious unauthenticated RCE risk via GraphQL.
OpenAI's Artifactory Enabled Covert Data Theft During Hugging Face Attack
Check Point Research uncovered a covert data theft channel in OpenAI's Artifactory.
Privacy Leakage in Split-LLM Training: A Security Case Study
A study on split-LLM training reveals privacy vulnerabilities in security checks.
Bobbin: Pentest with a Local LLM, Keeping Target Data Secure
Bobbin enables secure code reviews with local LLMs, ensuring sensitive data remains private during penetration testing.
One-Way Networks as a Solution to AI Security Risks
Exploring one-way network architectures and data diodes for AI security.