Clarence Reynolds, TelecomTV (00:08):
Sustainability and AI are poised to drive significant changes in the telecom industry, reshaping how networks operate and evolve. To address these pivotal topics, we are joined by Cristina Rodriguez, vice President of the Network and Edge Group, and General manager of the Wireless Access Network Division at Intel Corporation. Cristina, welcome. Before we dive into what to expect at this year's World Forum, I want to circle back to February where both at MWC and in our DSP Leaders Council Vision Report, we saw a key focus on sustainability. Specifically in the report, sustainability was key to respondents with 98% ranking it as either quiet or of the utmost importance. Can you tell us about what you're hearing in regard to sustainability and the work that Intel is doing to make networks more sustainable?

Cristina Rodriguez, Intel (01:04):
Clarence, definitely sustainability is top mind for all operators and really for the entire telco industry. We know that today telecom is one of the most powering dense industries on the planet. Operators are eager for solutions that will help them and help our industry reduce energy consumption, and this is essential for both reducing environmental impact but also reducing total cost of ownership. We have to deliver high performance, highly efficient networks, and one of the ways to do that is by having an end-to-end virtualized network. All the way to layer one. Operators need both the right architecture and the right silicone. Underpinning that architecture and focusing on solving this paradigm for our customers has guided Intel's vRAN roadmap investments and focus our collaboration with our partners. If you remember at NWC 2023, we introduced our fourth generation of Intel Cion SP called code name SFA Rapid EE with Intel VRA Boost, we call it Intel VRA Boost, fully integrated Acceleration, a first to market innovation.

(02:23):
This processor delivers twice the capacity with an additionally 20% power saving, and now these CPUs available and it's being deployed this year. And of course, we didn't stop there. We're committed to continue driving down vRAN cost and power and that this year, if you remember NWC 2024, we introduced the Future Gen Intel platform called named Cite Rapid D, and now this platform will deliver significant gains in performance and power efficiency, again, utilizing v REM boost acceleration and other architectural and feature enhancement, including built in AI acceleration. These platforms are built to deliver the right combination of performance, sustainability, TCO reliability. They're built basically for high performance, highly efficient AI optimized network. And just to give you an example, recently Verizon completed their first data call on our fourth generation C on SP with Intel V Boost, rapid E, and they highlighted, they say the following, they said they talk about the meaningful power efficiency gains, the ability to manage higher workloads and ability to manage higher throughput performance.

Clarence Reynolds, TelecomTV (03:52):
You mentioned artificial intelligence and another key focus in that report was AI, where about 80% of respondents believe that AI will drive a seismic shift in how telcos operate in the next five years. What do you think are the biggest opportunities with AI and ran?

Cristina Rodriguez, Intel (04:10):
Yeah, we know AI will be everywhere from client to edge to our network and cloud, and it will be accessible across all our loads, including CORE and ran. The potential uses for RAN AI are virtually unlimited locally. We have the network capabilities today to support AI in the RAN through software defined networks end to end, and we have CPUs that can run AI workload without requiring any external component, any additional power requirements like for example, the fourth generation, Intel four generation of C on scalable processor with Intel v Rambus. With that, you don't need anything else as far as opportunities. You were asking Clance, there are two major opportunities in Grant. One is about optimizing the ran, meaning optimizing performance and cost savings through AI and enhance automation, energy efficiency, dynamic orchestration. And the second major opportunity I'm going to say is creating new value added services that generate incremental revenue for operators.

(05:25):
For example, use cases that operators are more interested in today with immediate significant TCO benefits through performance and power improvements. Could be dynamic network configuration, traffic steering, optimize spectrum allocation, obviously energy efficiency in the infrastructure, predictive analytics and control automation in general. And then I believe that for revenue generation network slicing will play a key role and will be a major opportunities so far. Again, I'll give you some examples. We have multiple run AI demos with leading operators. We have a demo with SK Telecom that demonstrated AI assisting power savings. We have collaborated with Vodafone to demonstrate again, AI assisting network slicing. We have Deutsche Telecom, again, AI assisting B management, and we have worked with at and t again to demonstrate energies savings. So quite a bit happening already. We're just starting on this, but it has an enormous potential for both network efficiency and to enable new service models. And we have, again, we have the network capabilities and the CPU to start making happen that today.

Clarence Reynolds, TelecomTV (06:54):
So Cristina, as an emerging innovation, what is your advice to operators who want to streamline AI development and integration into their current network operations?

Cristina Rodriguez, Intel (07:05):
Yeah, as I said, we're just starting this journey, so it's not going to happen overnight. Operators will want to make sure the algorithm works. They can deploy safely that they're not going to have any compromise. It's just going to get better and better, and that's what they're doing. There is already elements of AI being incorporated in today's network. I think again, we start by having a software defined network. We embrace the new architecture software defined network end-to-end virtualize where innovation can happen at the pace of software, and you choose an architecture with capabilities that can make it easy and practical to deploy. Having that architecture silicon underpinning that architecture that allows you to deploy AI without adding additional components and without adding power consumption, reducing the complexity and that reducing the cost, that is key. Every single AI use case inference use case that can be applied to the virtual DU can be run on Intel C processes, and that is huge when not in the data center or at the border of the tower or a building, TCO is essential performance per what still drives architecture and business decisions, and we have to take that to keep that in mind when we choose the solutions.

(08:36):
Also, another thing to keep in mind is software migration, right? That we get when you architecture is running on a Intel C on server architecture, you get software migration, you get full compatibility generation over generation, and that protects the investment, the software investment that you will be doing today. I want to mention this, another intel contribution to help operators early in their AI journey is what we just announced again at the WC 2024, the early availability of the Intel vRAN AI development kit. It enables operators and developers to build models to build, to train, to optimize and deploy AI models for vRAN use cases built on top of Intel AI optimized libraries, frameworks and tools. So really a very, very useful tool.

Clarence Reynolds, TelecomTV (09:41):
Excellent. And I'd like to transition now to the world forum where our theme this year represents what's at the top of people's mind here in this telecom industry. How can we unleash the digital services opportunity? That's the question that we asked, and from your point of view, Cristina, how can we accelerate network evolution?

Cristina Rodriguez, Intel (10:01):
We're seeing what is happening in the telco world. We're seeing a massive industry transformation toward a software defined end-to-end virtualized network, all the way to layer one in the entire stack. We're seeing major operators across the entire world either already deploying at a scale, starting to deploy, or definitely announcing their intention to deploy in both Greenfield and Brownfield. We're seeing Vodafone, we're seeing Verizon, we're seeing the recent announcements from at and t. We're seeing Tes Telefonica Dish, Rakuten in Japan, so many others. We have reached the point of not return. That's just a fact, and we can do this. We know now that we can do this transformation without compromising the KPIs. We have seen the reports from Verizon, from Vodafone, from Rakuten on their KPIs on their success with their deployment, the deployment of the network. The industry has embraced an open virtualized network that can bring scalability, flexibility, innovation in multiple areas, including AI handstand without again, compromising KPIs. I'm very proud that today nearly every virtual open brand deployment in the world runs on Intel Xeon. We led virtualization at the core, if you remember that, and our leading virtualization at the brand. From our viewpoint, we're continuing to focus on advancing the technology, investing our silicon platform and software roadmaps to simplify and accelerate operator deployment. Yet unleashing the full opportunity of the virtual open network will require collective innovation and collaboration across the industry, which is why efforts like the DSP Council are so critical to transforming our network.

Clarence Reynolds, TelecomTV (12:01):
And earlier you mentioned the critical role of ecosystem collaboration and the success of 5G and beyond. At MWC, we saw the wide breadth of Intel's partnerships in the industry. Can you tell us how these collaborations contribute to advancing network innovation and scalability in the years to come?

Cristina Rodriguez, Intel (12:22):
Absolutely. This is a task for the whole industry. No one can do this alone as an ecosystem. We need to work together to simplify and accelerate virtual open run deployments and bring innovation at the pace of software. We'll continue to work with our partners to optimize the hardware and the software together, and we're lucky and thankful for the amazing and rich ecosystem. At anos. We'll continue to provide ecosystem with solutions that give the operators the best results. Scalability, flexibility, innovation, including AI workloads, power reduction automation, workload consolidation. It's unlimited. The more we can collaborate upfront together, for example, aligning our roadmaps, optimizing and validating our solutions upfront, the more we can simplify operators, virtual, open, run deployments, and the faster we can scale and give operators the best results.

Clarence Reynolds, TelecomTV (13:25):
Cristina, thank you so much for your insights today.

Cristina Rodriguez, Intel (13:28):
Thank you, Clarence. Always great talking to you.