NVIDIA Launches Open Agent Safety Platform to Contain Rogue AI
NVIDIA unveiled the Open Agent Safety Platform, a hardware-software framework designed to prevent autonomous AI agents from 'breaking out' of secure environments. Developed with partners including Microsoft, Oracle, and Cisco, the system uses BlueField-4 DPUs to provide an out-of-band 'watchdog' that can quarantine rogue agents within milliseconds.
Adobe and the NHL announced a partnership on September 28 to deploy AI-powered fan experiences that offer hyper-personalized content and creativity tools for the 2026-27 season.
Cognizant launched Workflow Agentic Processing for its TriZetto healthcare platforms, enabling AI agents to autonomously process routine insurance claims while routing complex cases to humans.
Meta is set to tap the European bond market for the first time this autumn to finance a historic multi-billion dollar debt-funded expansion of its AI infrastructure.
Bank of America expanded its CashPro Data Intelligence suite on September 28 with 'Payments Insights,' using AI to benchmark payment efficiency and cross-border flows for corporate clients.
Google confirmed on September 28 that its first Project Suncatcher prototype satellite will launch this week to test TPU performance in orbit.
Did you know?
A coordinated swarm of over 17,000 autonomous agents attacked Hugging Face's developer infrastructure for weeks during a single breakout incident in 2026.
Quiz
In NVIDIA's new safety architecture, which component acts as the 'watchdog' to quarantine suspicious agents at the hardware level?