Nvidia and Nokia expand partnership to build sovereign AI for telecoms

Automating the network

Nvidia and Nokia have announced an expansion of their partnership to develop a sovereign agentic AI platform. This technology is designed specifically for telecommunications operators who need to automate their network operations without sacrificing control over their own infrastructure.

The goal is to create a system where AI agents—software programs capable of perceiving their environment and taking actions to achieve goals—manage the heavy lifting of network maintenance. These agents will monitor various parts of the network, including the radio, core, transport, and cloud layers. By identifying issues and taking corrective actions automatically, the system aims to reduce the amount of manual work required by human engineers.

As Mobile Europe reported, the platform integrates Nvidia AI Enterprise software and Nemotron models with Nokia's established network technology. This combination allows operators to build and deploy these agents across their existing systems to improve overall performance and reliability.

Data sovereignty and control

A major hurdle for many large enterprises, especially in telecommunications, is the risk of sharing sensitive operational data with external providers. To address this, the new platform emphasizes 'sovereign' AI. This means that operators maintain full ownership and oversight of their data, models, and specific policies.

The partnership offers three different deployment options: on-premises, cloud, or hybrid. This flexibility allows a company to decide exactly where their information is stored and processed. For example, a government-linked telecom might prefer an on-premises setup to keep all data within their own physical buildings, while a smaller provider might opt for a cloud-based solution.

Nokia stated that the system includes specific safeguards. These features are intended to let human operators oversee how the AI agents make decisions and carry out their assigned tasks. This creates a balance between the speed of automation and the necessity of human supervision in critical infrastructure.

The path to 6G

While the technology is being applied to current 5G networks, the partnership is also looking toward the future of connectivity. Nvidia and Nokia are positioning this software-defined architecture as a foundation for 6G networks. These next-generation networks are expected to be 'AI-native,' meaning they will be built from the ground up to distribute intelligence rather than just moving data from point A to point B.

Nvidia Senior Vice President of Telecom Ronnie Vasishta noted that the companies are combining CUDA accelerated computing with Nokia's expertise to improve radio performance. This infrastructure is intended to support 'physical AI,' a concept where AI interacts with and manages physical systems in the real world. By making the network more autonomous, operators can get more value out of their existing spectrum while preparing for new types of AI-powered services.

The companies are building on their existing AI-RAN partnership. This initiative is already being tested by several major global operators, including Orange, BT, Deutsche Telekom, and Elisa. The expansion with Nokia and Nvidia aims to turn these tests into a practical, wide-scale reality for the telecom industry.

Who is using it now

The technology is not just a theoretical project; several major telecommunications companies are already involved in exploring its applications. Nokia is currently working with etisalat UAE (formerly e!), MEO, and Airtel to see how these AI agents can be best utilized in their specific network environments.

For these operators, the transition to agentic AI represents a shift in how they manage hardware. Instead of reacting to a network failure after it happens, the AI agents are designed to proactively monitor conditions and optimize the network in real-time. This move toward a software-defined foundation is intended to make networks more resilient and capable of supporting the increasing demands of modern data usage.

Technical foundations

To understand how this works, it helps to look at the underlying components. Nvidia provides the computational 'brain' through its Nemotron models and AI Enterprise software. These provide the reasoning capabilities that allow the AI to understand complex network patterns.

Nokia provides the 'nervous system' of the telecom world. Their network software and expertise in radio and transport layers provide the framework where the AI agents live. By combining these, the two companies are creating a system where the AI can 'see' the network status and 'act' on it using Nokia's established protocols. This integration is what allows for a trusted environment where the AI can operate autonomously without creating security vulnerabilities.

Next steps for the industry

The expansion of this partnership signals a broader trend in the tech industry toward sovereign AI. As more companies look to adopt AI, the ability to do so without moving data to a public cloud is becoming a primary requirement for many industries, particularly those involving national infrastructure.

Moving forward, the success of this platform will depend on how well the AI agents can handle unpredictable network stresses. Operators will be looking for evidence that these agents can maintain high performance during peak usage times and successfully mitigate cyber threats. As the technology matures, we can expect to see more operators moving away from manual configuration toward a model where AI agents handle the majority of routine network maintenance.

Looking ahead

Nokia Chief Technology and AI Officer Pallavi Mahajan highlighted that the next era of connectivity will be defined by networks that distribute intelligence. This means that instead of a central hub making every decision, the intelligence will be spread across the network, allowing for faster and more efficient performance.

The partnership aims to create a practical path toward increasingly autonomous networks. By providing a software-defined foundation, Nvidia and Nokia are making it easier for telecom companies to adopt AI-native features. This will likely lead to new types of services for consumers, ranging from more reliable mobile connections to new ways for physical devices to interact with the internet of things.

The role of CUDA

A key part of Nvidia's contribution to this partnership is CUDA. CUDA is a parallel computing platform and programming model that allows developers to use Nvidia GPUs (Graphics Processing Units) for general-purpose processing. In this context, it provides the accelerated computing power needed to run complex AI models in real-time across a massive telecom network.

By using CUDA-accelerated computing, the AI agents can process vast amounts of data from millions of devices simultaneously. This speed is essential for maintaining the low latency required for 5G and future 6G services. When combined with Nokia's network software, it creates a high-performance environment where the AI can make split-second decisions to keep the network running smoothly.

Impact on 6G development

The move toward 6G is one of the primary drivers for this partnership. 6G is expected to offer much higher speeds and lower latency than 5G, but it will also require much more complex management. The industry believes that human operators will not be able to manage the scale of 6G networks manually.

By establishing a sovereign AI platform now, Nvidia and Nokia are ensuring that the infrastructure for 6G is ready to be managed by AI from day one. This proactive approach allows telecom companies to build a foundation that is both intelligent and secure. It paves the way for a future where networks can automatically adjust to user demand, repair themselves, and provide the backbone for the next generation of AI-powered applications.

Summary of the partnership

In summary, the expanded partnership between Nvidia and Nokia is a strategic move to bring sovereign AI to the telecommunications sector. By combining Nvidia's AI models and computing power with Nokia's network expertise, they are creating a tool that allows operators to automate their networks while keeping their data private.

The platform supports a variety of deployment styles, ensuring that different types of companies can use the technology regardless of their existing infrastructure. With major operators already testing the system, the move toward autonomous, AI-native networks is well underway. This collaboration marks a significant step in how the world's connectivity will be built, managed, and optimized in the coming years.

Key takeaways

The core of this announcement is the shift from manual network management to autonomous, AI-driven operations. For the telecom industry, this means less downtime and more efficient use of their spectrum. For the broader tech world, it demonstrates how sovereign AI can be applied to critical infrastructure.

The partnership also highlights the importance of software-defined architecture. By making the network software-defined, it becomes much easier to integrate AI agents that can be updated and scaled quickly. This flexibility is what will allow the industry to move toward 6G and beyond, providing a trusted foundation for the next generation of digital services.

Final thoughts

As the technology moves from testing to wider deployment, the focus will likely shift to the specific use cases that operators find most valuable. Whether it is optimizing radio performance or identifying core network bottlenecks, the goal remains the same: making networks smarter and more efficient through the power of sovereign AI.

Nvidia and Nokia are not just providing a new tool; they are helping to define the architecture of the future internet. By creating a system that balances automation with human oversight, they are providing a practical roadmap for how the world will stay connected in an increasingly AI-driven era.

Technical details

The platform relies on a specific stack of technologies to achieve its goals. Nvidia's Nemotron models are the underlying AI models that provide the reasoning and processing power. These models are trained to handle complex tasks and can be fine-tuned for specific telecom needs.

Nokia's network technology provides the actual pipes and switches that carry data. By integrating these two, the AI agents can interact with the physical reality of the network. The result is a system that can 'understand' a drop in signal and 'decide' to reroute traffic or increase power to a specific radio tower, all without a human having to type in a command.

The significance of 'Sovereign'

The word 'sovereign' is key in this announcement. In the context of AI, it refers to the ability of an organization to maintain complete control over its data, models, and the environment in which the AI operates. This is a major concern for many countries and large corporations who are wary of sending sensitive information to public cloud providers.

By offering on-premises and hybrid options, Nvidia and Nokia are catering to this need for data residency. This ensures that a telecom company's data stays within its own borders and under its own security protocols. It allows for the benefits of AI—like automation and optimization—without the risks associated with centralized, third-party AI systems.

Operator involvement

The involvement of major operators like etisalat UAE, MEO, and Airtel is a strong indicator of the technology's commercial viability. These companies are at the forefront of the telecom industry and have the scale to test these systems in diverse environments.

Their participation suggests that the industry is ready to move toward AI-native networks. As these operators gather data on how the AI agents perform, they will be able to refine the platform to meet the specific needs of different regions and types of network traffic. This collaborative approach ensures that the final product is practical and effective for real-world use.

Future of 6G

The roadmap toward 6G is a central theme of this partnership. 6G is expected to be much more than just a faster 5G; it is envisioned as a platform for 'physical AI.' This means AI will be integrated into the very fabric of the network to manage everything from autonomous vehicles to smart cities.

By building the foundation today with sovereign AI, Nvidia and Nokia are ensuring that the infrastructure is ready for these future demands. The transition to 6G will require networks that can self-optimize and distribute intelligence across vast distances. This partnership provides a clear path toward achieving that goal while maintaining the security and control that modern society requires.

Software-defined architecture

A software-defined architecture is a way of designing networks where the functions are separated from the underlying hardware. This allows the network to be more flexible and easier to manage through software updates rather than physical changes to the equipment.

This is a critical component of the Nvidia and Nokia partnership. By using a software-defined approach, the AI agents can be deployed more easily across different types of hardware. It also allows for the 'trusted, software-defined foundation' that Nokia mentioned, where security and performance can be managed through code. This is the standard for modern telecommunications and is essential for the deployment of AI at scale.

Conclusion

The expansion of the Nvidia and Nokia partnership is a significant milestone in the evolution of telecom networks. By combining AI expertise with network infrastructure, they are creating a system that can automate complex tasks while keeping data secure. This move toward sovereign AI is a key step in building the autonomous networks of the future.

As operators begin to deploy these AI agents, we can expect to see more reliable and efficient connectivity. The technology provides a practical way for the telecom industry to embrace AI, ensuring that the next generation of networks is not only faster but also smarter and more capable of supporting the needs of a digital world.