Hacker News Reader

When you say firewalls for AI models, you mean firewalls for agents, right? Otherwise, I don't see the need to firewall a model.

View on HN

I meant for a model. Like agents could be many within an organization, instead of patching up each agent with a security solution, wouldn’t it be easier to build one that wraps around the model ?

Reply on news.ycombinator.com