NVIDIA announced today a strategic collaboration with Google Cloud to deploy Google’s Gemini family of AI models on-premises using NVIDIA’s Blackwell platform and Confidential Computing technology.
This partnership specifically targets organizations that need agentic AI capabilities but face strict regulatory requirements and data sovereignty constraints.
The Agentic AI Advantage
Unlike traditional AI systems, agentic AI offers advanced reasoning capabilities that transform how enterprises solve complex problems. NVIDIA announced:
Unlike AI models that perceive or generate based on learned knowledge, agentic AI systems can reason, adapt and make decisions in dynamic environments.
These capabilities unlock powerful new applications across industries.
In enterprise IT support, agentic AI systems can take automation to a new level by moving beyond basic information retrieval. Instead of simply pulling up troubleshooting guides, these reasoning-based systems can actively diagnose technical issues, implement fixes, and know when to escalate more complex problems to human specialists.
This capability extends to financial operations as well, where agentic AI can surpass traditional pattern-matching to actively investigate suspicious activities and implement protective measures—like blocking potentially fraudulent transactions or dynamically updating security rules—all without waiting for human intervention.
The On-Premises Solution
The collaboration brings Google’s Gemini models to on-premises environments through Google Distributed Cloud running on NVIDIA Blackwell HGX and DGX platforms. This addresses what NVIDIA calls “the on-premises dilemma” – until now, organizations with stringent security requirements couldn’t access the full capabilities of agentic AI from leading labs due to cloud-only deployment models.
Sachin Gupta, vice president and general manager of infrastructure and solutions at Google Cloud, said:
By bringing our Gemini models on premises with NVIDIA Blackwell’s breakthrough performance and confidential computing capabilities, we’re enabling enterprises to unlock the full potential of agentic AI. This collaboration helps ensure customers can innovate securely without compromising on performance or operational ease.
Security and Scale: The Foundation for Enterprise AI
NVIDIA’s Confidential Computing technology establishes the security cornerstone of this partnership. It creates protected enclaves for both API prompts and training data, ensuring patient records, financial transactions, and classified information remain locked down from unauthorized access. This dual-layer protection directly addresses the privacy concerns that have kept regulated industries from adopting advanced AI.
The solution extends beyond just security. Google Cloud’s new GKE Inference Gateway works with NVIDIA Triton Inference Server and NeMo Guardrails to intelligently route and balance AI workloads, cutting costs while maintaining central governance. Future plans include integrating NVIDIA Dynamo to handle the distinct challenges of running reasoning models at enterprise scale.
Together, these technologies tackle both the security and operational hurdles that have prevented many organizations from deploying sophisticated AI systems. For the first time, companies can run powerful reasoning models behind their own firewalls without sacrificing performance or compliance.
Real-World Impact and Market Positioning
Healthcare systems can now analyze medical records while keeping patient data behind their own firewalls. Banks can deploy fraud detection systems that proactively adjust security measures without exposing customer information. Government agencies can process classified information with necessary security clearances intact. The Gemini models’ ability to process text, images, and code simultaneously tackles the complex, multi-format challenges these organizations face daily.
This partnership arrives amid strategic moves from both companies. Google recently launched Gemini 2.5 Pro, enhancing its reasoning capabilities, while NVIDIA continues expanding its AI hardware roadmap despite facing market volatility earlier this year. The collaboration extends their existing joint work across robotics, energy grid optimization, and drug discovery.
By becoming one of the first providers offering secured agentic AI workloads across cloud and on-premises environments, this solution opens advanced reasoning capabilities to organizations previously locked out by security constraints. The combined approach addresses both technical performance and regulatory compliance—two barriers that have kept many from fully embracing AI’s most powerful applications.
