Nokia and NVIDIA collaboration accelerates AI-RAN deployment across global operators

Added: 2026-05-05

[00:00:03][music] >> Hello, you're watching Telecom TV. I'm Guy Daniels.

[00:00:16]The radio access network is undergoing a profound transformation with the integration of AI, and this is set to accelerate as the capabilities of AI prove themselves.

[00:00:28]To help us understand the shift to AI RAN and what it means for telcos, I'm joined today by Ajit Ed, who is VP, head of AI RAN and Cloud RAN at Nokia. Hello Ajit, it's good to see you again. Um, can I just start by asking you really, you know, straightforward question here, but we need to get this clear. What is the real difference between RAN and AI RAN at an architecture level? What makes it different?

[00:00:57]Thank you, Guy.

[00:00:59]So, if you look at the the traditional RAN, it's [clears throat] all about connectivity.

[00:01:04]It's largely, if you look at it, fixed functions, hardware-centric, and designed around relatively predictable traffic patterns.

[00:01:12]But when it comes to AI RAN, it it is really a structural shift. It's software-defined, highly programmable, accelerated computing, and if you look at it from a designed from outset to support both RAN workloads and AI workloads from the beginning.

[00:01:30]The key distinction here is whether it is AI native or it's an AI augmented.

[00:01:36]When is When I say AI native, that means the compute, connectivity, control all are co-designed, designed from the outset, designed from the beginning.

[00:01:46]So, AI is not simply added on top of RAN.

[00:01:50]RAN itself becomes programmable compute platform. And this is the key difference, and this is essential for us to proceed into AI native 6G network.

[00:02:00]Great, that's very clear. So, there's a structural shift going on here.

[00:02:04]What are the benefits then of AI RAN for a telco because there has to be a compelling reason for them to make the change?

[00:02:11]Of course. I mean, of course, the the the key the topics about AI RAN are moving into three different buckets. If you look at the AI for RAN, AI on RAN, and AI end RAN. And AI for RAN, as the name implies, it's all about improving the efficiency of the network, improving the spectral efficiency. And how does it help with the GPU accelerated computing?

[00:02:34]Yes, it helps in multiple different ways. Number one, if you look at the parallel computing capabilities of GPU, and GPUs are meant inherently for the parallel computing, and there are different algorithms what we can think of, beamforming at once, beamforming, etc., which will have lot of parallel computing possibilities which are helped by the GPUs.

[00:02:57]And second, we can also bring in machine learning model-based uh algorithms, channel estimations, uh multi-user MIMO pairing, and many more, carrier aggregation, and and so on. So, these all feature sets, the RAN functions are helped by machine learning models, and these are inherently helped by the accelerated computing. And the third piece, it's about bringing new algorithms on top of the existing ones.

[00:03:24]Like, for instance, AKHS, and this is a compute-intensive uh algorithm, and we can make use of GPUs apart at GPU accelerated computing for bringing such complex algorithm and getting the benefits for the RAN efficiency. And this is in the AI for RAN.

[00:03:42]And AI on RAN, it's about bringing the new use cases on top of RAN. For instance, uh we talk about physical AI, we talk about Isaac, intelligent sensing and uh communication. These are some of the use cases where the compute capabilities of at accelerated computing are used you for these use cases and better [clears throat] suited for that.

[00:04:05]And the third bucket is AI end RAN, and this becomes a uh AI architecture AI RAN network becomes a multi-purpose cloud platform, multi-purpose platform where AI workloads and RAN workloads can coexist and uh work together in using the same infrastructure. It opens up a lot of new revenue opportunities for uh operators.

[00:04:31]So, as you say there, there's a lot of new revenue opportunities for operators.

[00:04:35]Can you go into some more detail now about the new business and monetization models that AI RAN is creating? So, if you look back, so in case of 2G and 2G a few decades ago, it started with um voice, a little bit of data, and then when 4G came in, it's more about data consumption, and 5G became more video, video streaming applications, and now if you look at it, AI workload becomes a fourth workloads of the telco network.

[00:05:05]Networks are becoming the critical infrastructure for AI.

[00:05:10]And as you can imagine, AI workloads are highly sensitive to uh jitter, latency, and the worst-case performance uh the nature of the network.

[00:05:20]And at the same time, the traffic patterns are increasingly more dynamic and more upling intensive. And all these enterprise customers expect the performance guarantees, the SLAs, automation, seamless cloud migrations, and all of these pieces.

[00:05:38]So, this AI RAN allows the operators to leverage their biggest asset. They have the distributed cell sites, distributed computing infrastructure what they have today.

[00:05:49]And we call it as we can call it as AI grid.

[00:05:53]And this will enable with the accelerated computing on distributed cell site and MSOs switching centers, this enable the edge inferencing and token processing at the network edge.

[00:06:06]This really creates a monetization opportunities beyond connectivity, AI services, edge AI services, uh enterprise automation use cases, SLA-based enterprise offering, and many more.

[00:06:19]So, here all of these cases, this AI RAN is going to be a catalyst for providing new opportunities for operators.

[00:06:27]So, this is a really interesting new use case here. So, can you give us some specific AI use cases that actually benefit from edge processing?

[00:06:37]Yeah, of course. I mean, if you look at the the kind of AI traffic now, the scale of AI traffic, it's really enormous. It already justifies the edge processing now.

[00:06:49]I mean, if you look at the overall amount of AI interactions happening annual basis, it's more than trillion.

[00:06:56]And with all these leading platforms, they're generating tens of trillions of tokens per day.

[00:07:02]This significant share of generative AI usage is coming from already from the mobile.

[00:07:09]So, if you look at the different segments of AI use cases, generative AI or agentic AI, of course, edge processing helps to reduce the latency.

[00:07:18]But more importantly, this helps to offload uh the AI processing from the centralized cloud data centers to the edge.

[00:07:29]And if you look at the physical AI, that's including the robotics, drones, and real-time machine control, so edge processing is an essential.

[00:07:39]Right, because it's extremely critical to have a less jitter, lower latency, deterministic latency, and extremely reliable bidirectional communication.

[00:07:50]And all of these are enabled by edge inferencing either at the distributed cell site or at the MSO switching offices. So, it can always start from the switching office inferencing, which is closer to the user. Then, if this is there is a need, of course, this platform, the beauty of this platform such that this will enable such inferencing capabilities already at this distributed cell site, uh which can happen anytime in the future.

[00:08:16]Well, Ajit, we know that AI developments move extremely fast. What's the latest progress on AI RAN, and what market momentum are you now seeing?

[00:08:26]Yeah, we are we are seeing a tremendous momentum with the AI RAN engagement with our customers, partners all across the world. When we launched the AI RAN collaboration with Nvidia back in October 2025, since then we have been working extremely closely with Nvidia to really bring the new solutions. And if you see the the events like MWC this year and the GTC uh in San Jose in March, they demonstrated the real AI RAN uh validation of this platform, and it's moving from the concept to execution.

[00:09:04]And we have the earliest customer trials, the ecosystem alignments are already happening. So, this is really driving the evolution of 5G towards the creating a foundation for AI native 6G.

[00:09:16]So, the momentum across the global operators, again, it's again significant. We announced more than 10 new collaborations [clears throat] with the operators around the world during MWC this year, including uh British Telecom, um NTT Docomo in Japan, Elisa in Finland, Vodafone Group. This is all in addition to the the the real leading partners like the T-Mobile USA, SoftBank in Japan, and Ooredoo Hutchison Indosat in Indonesia. So, this leveraging the AI aerial platform from Nvidia and using our any run software built on top of it.

[00:09:55]And it's at the same time, we're also expanding the ecosystem partners.

[00:10:00]Because we strongly believe the ecosystem is extremely crucial. And we are expanding the partners to like the likes of Quanta, Supermicro in the server infrastructure side, and also on the CAS side with the Red Hat. So, the momentum is great. We are now focused on the execution of AI RAN, and we have clear plans in terms of bringing all these pieces together to our operators this year and next year in 2027. That's really good to hear. And I've got a final question for you. I'd love to know how the relationship is developing between Nokia and Nvidia, and how both companies are benefiting from working together.

[00:10:37]Yeah, so if you look at it, so we bring in two leading powerhouses.

[00:10:43]And Nvidia is an AI powerhouse, and and Nokia is a telecom powerhouse. So, these two leading companies are coming together to change the AI network architecture for the telcos.

[00:10:56]So, our collaboration with Nvidia is um advancing rapidly, as I already talked about.

[00:11:02]This is all about ecosystem creation.

[00:11:05]It's all about co-creation with our customers.

[00:11:07]And because we believe that it's not realized AI RAN is not realized by a single company. It's an ecosystem what is needed. So, together with the Nvidia, we created an AI RAN platform built on top of the Ampere compute platform and using our any RAN software. And we have already successfully validated any RAN software on top of this platform.

[00:11:29]So, this enables the faster innovation cycle. We can bring in the software features much faster than it was before.

[00:11:37]It is all based on a software-driven architecture.

[00:11:40]And it reduces the integration risk for our customers.

[00:11:44]And it creates a foundation for a broader ecosystem of developers and application providers, and which they can use it to innovate further.

[00:11:55]So, for operators, it means that they can stay ahead in this AI super cycle with a faster innovation cycles, rather than just reacting to it.

[00:12:06]That's great to hear. We must leave it there, actually good talking with you, as always. And thanks very much for sharing your views with us today. Thank you, Guy.