The swift emergence of “open-clos,” autonomous AI agents adept at drafting reports and executing tasks independently, faces escalating scrutiny, particularly in China, a major user. Authorities have opted to restrict these systems in government and state-owned enterprises due to security concerns over potential data leaks and unauthorized access to sensitive information. Notable incidents, such as a Meta-developed AI exposing data, underscore these risks. Meanwhile, Nvidia has launched “NeMo-Clos,” presenting a secure alternative that emphasizes privacy protection and oversight. Open-clos, developed by Peter Steinberger, operates beyond traditional chatbots, executing complex tasks autonomously. Rapidly adopted in China—with 1,000 participants at a Tencent event—the tool has drawn attention for its operational capabilities. However, mounting concerns were voiced by regulatory bodies, warning of potential loss of control and system breaches. In response to the security landscape, global tech companies like Nvidia and Alibaba are racing to innovate enterprise-grade AI platforms with fortified safeguards.
Source link
Share
Read more