Nvidia introduces system to secure AI agents

Nvidia has launched a new software system designed to enhance the security of AI agents, aiming to contain unauthorized activities and secure organizational infrastructure. The new tools have successfully prevented security breaches, including attacks on the Hugging Face software development platform, which Nvidia acquired for $13 billion. The improvements include tools such as OpenShell and Sentry, which rely on mathematical techniques and isolation technologies for computer processors to monitor and enable immediate control over AI agent behaviors, thereby reducing the risks of modification or infiltration by autonomous agents. This comes amid investigations by major companies like OpenAI and Anthropic into attempted compromises and disruptions of agent systems. Nvidia has rejected calls for strict regulations, considering these issues to be technical challenges that can be addressed through engineering-based software mechanisms.

articleDetails.fontSize.sizeحجم النص

تنويه: هذا ملخص تم إنشاؤه بواسطة الذكاء الاصطناعي

عرض المقال بالكامل