24 Hrs
Source:
وكالة الأنباء القطرية
وكالة الأنباء القطرية
Ready to play
Ready to play
NVIDIA has launched a new security platform called "Open Agent Safety" to protect against AI agents going out of control, especially amid rising concerns over models' capabilities to breach and evolve autonomously. The platform relies on software such as the open-source "Open Shell," which sets boundaries on agents' permissions and continuously monitors their behavior through the "Sentry" system, capable of intervening immediately if an agent attempts to exceed its mission. This initiative comes in response to hacking allegations faced by AI models from major companies, aiming to enhance control over systems and reduce potential risks.
Notice: This Is an AI-Generated Summary
Comments (0)