Open research, controls across the agent stack and continuous testing help defenders build and operate more secure AI systems.

3 points•ketanbj•2 days ago•1 comment•

1 comment

ketanbj2 days ago
I think this is a step in the right direction. However, there seems to be too many encumbrances and too much reliance of existing technology to be sufficient to address AI agent safety.

We at ArmorIQ welcome it and think there is more to it then meets the eye. https://armoriq.substack.com/p/nvidia-is-moving-agent-safety...

Read the full thread on Hacker News →

Related stories