Multimodal-Safety | GRID THE GREY

LIVE FEED

HIGH AI Guardrails Fail Multilingual Jailbreak Tests in Europe // FIRST LOOK AI Agent Security Shifts From Visibility to Enforcement Controls // HIGH Hermes AI Agent Automates Post-Exploitation Attack on Thai Finance Ministry // FIRST LOOK Google Gemma Tech Brings 28.9M LLM to ESP32 Microcontrollers // HIGH AI Coding Agents Exploited via Hallucinated Package Names // FIRST LOOK AWS Adds Bedrock Guardrails Best Practices for Code Generation // FIRST LOOK Anthropic Brings Voice Mode to Claude Opus and Sonnet Models // FIRST LOOK Anthropic and OpenAI Open Vetted Cyber Programs for Offensive Researchers // HIGH Threat Actor Trim Weaponises AI Jailbreaks for Offensive Ops // HIGH Fake Claude App via Bing Ads Delivers SectopRAT Malware //

1 report

All LLM Security Agentic AI Industry News Research Supply Chain Prompt Injection First Look: Security Regulatory Adversarial ML Jailbreaks Model Theft Data Poisoning AI Security Tools Security Operations

Llama Guard 4 Jailbreak Detection Vulnerable to Prompt Injection

Llama Guard 4 Jailbreak Detection Vulnerable to Prompt Injection

ATLAS OWASP LOW ▲ 7.2 Hugging Face Blog Apr 28, 2026

Meta has released Llama Guard 4, a 12B multimodal safety classifier designed to detect and filter unsafe content in both image and text inputs/outputs for production LLM deployments. The model addresses jailbreak attempts and harmful content generation across 14 hazard categories defined by the MLCommons taxonomy. Alongside it, two lightweight Llama Prompt Guard 2 classifiers (86M and 22M parameters) target prompt injection and prompt attack detection.