Nvidia's Open Agent Safety Platform: OpenShell Sandboxes, Sentry Watchdog Take Aim at Rogue Agents
Posted on 28th Sep 2026 12:05:31 in Artificial Intelligence, Machine Learning
Tagged as: nvidia, ai agents, agent safety, ai security, openshell, sentry, agentic ai
Nvidia launched the Open Agent Safety Platform — OpenShell sandboxes in software, a Sentry watchdog in BlueField-4 silicon — to stop AI agents from breaking out of containment after a month of rogue-agent incidents at OpenAI, Google and beyond.
OpenAI Pauses Astra Work as Model Approaches 'Critical' Cybersecurity Threshold
Posted on 9th Aug 2026 07:41:17 in Artificial Intelligence, Machine Learning
Tagged as: OpenAI, Astra, AI Security, Cybersecurity, AI Safety, Artificial Intelligence, LLM
OpenAI says it cannot rule out critical cybersecurity capabilities in its upcoming Astra model, triggering stricter safeguards, isolated testing, and a development pause under its Preparedness Framework.
Microsoft Launches MAI-Cyber-1-Flash: First In-House AI Cyber Model Tops CyberGym Benchmark
Posted on 31st Jul 2026 06:03:36 in Artificial Intelligence, Machine Learning
Tagged as: Microsoft, AI Security, Cybersecurity, Agentic AI, MDASH
Microsoft unveiled MAI-Cyber-1-Flash, its first in-house AI cyber defense model, running inside the MDASH multi-agent harness — scoring 96% on CyberGym at half the cost.
