Rebeca Moen
April 11, 2025 17:09
NVIDIA has a partner relationship with Google Cloud, introducing Agency AI for enterprises through the BLACKWELL architecture and confidential computing, and improves data security and performance.
NVIDIA worked with Google Cloud to provide enterprises with agent AI features using the Google Gemini model of the NVIDIA BLACKWELL infrastructure. According to NVIDIA, this collaboration aims to improve the ability of companies that utilize AI while maintaining data security and compliance.
Improvement of security and performance
Companies will be integrated with Google Distributed Cloud to benefit from NVIDIA’s BLACKWELL Architecture, which guarantees data sovereignty and regulatory compliance. NVIDIA Confidential Computing Framework further secures sensitive data to protect it from unauthorized access and potential leaks. Through this dual protection, companies can ensure data personal information by safely innovating with the Gemini model.
Sachin GUPTA, vice president of infrastructure and solutions of Google Cloud, emphasized the importance of this collaboration that enables safety innovation without damaging performance. Integrating NVIDIA’s infrastructure and twin seat models is set to unlock the potential of agent AI for business applications.
The possibility of deformation of agent AI
Agency AI is a revolutionary to enterprise technology by providing advanced problem solving functions. Unlike existing AI models that depend on learned knowledge, agent AI systems can autonomously reason, adapt and decisions. This ability is especially valuable in areas such as IT support and finance, which can diagnose problems, execute modifications, and actively prevent fraudulent activities.
Solve the on premises requirements
Many companies are having a hard time distributing AI model premises due to strict security or data sovereignty requirements. The new solutions of NVIDIA and Google Cloud solve this problem by providing a confidential computing feature that protects agent AI workloads in all environments such as cloud -based or hybrids.
The NVIDIA HGX B200 platform, driven by the Blackwell GPU, improves energy efficiency and performance while protecting the AI model and data integrity. This development allows companies to use cutting -edge AI without sacrificing security.
AI observation potential and security improvement
Effectively expanding agent AI requires powerful observation and security measures. Google Cloud introduces a new GKE reasoning gateway to optimize the AI reasoning workload through advanced routing and scalability. Integrate with the NVIDIA Triton reasoning server and Nemo Guardrails to ensure the central security and governance of the AI model.
Google Cloud will integrate NVIDIA Dynamo, an open source library designed to expand the AI model, to improve the possibility of observation of agent AI workloads. This initiative will support additional companies that efficiently distribute and manage AI applications.
Image Source: Shutter Stock