<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Net AI on Medium]]></title>
        <description><![CDATA[Stories by Net AI on Medium]]></description>
        <link>https://medium.com/@netai?source=rss-a2d8f7b90894------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*t6vHc7CR1TTD5MGHxTzPKg.png</url>
            <title>Stories by Net AI on Medium</title>
            <link>https://medium.com/@netai?source=rss-a2d8f7b90894------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Wed, 18 Nov 2026 01:24:39 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@netai/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Net AI selected by Digital Catapult to support trusted Telecom AI adoption in national critical…]]></title>
            <link>https://medium.com/net-ai-insights/net-ai-selected-by-digital-catapult-to-support-trusted-telecom-ai-adoption-in-national-critical-fdc16c29dc12?source=rss-a2d8f7b90894------2</link>
            <guid isPermaLink="false">https://medium.com/p/fdc16c29dc12</guid>
            <category><![CDATA[network-data]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[critical-infrastructure]]></category>
            <category><![CDATA[energy-efficiency]]></category>
            <dc:creator><![CDATA[Net AI]]></dc:creator>
            <pubDate>Mon, 05 Oct 2026 08:58:09 GMT</pubDate>
            <atom:updated>2026-10-05T08:58:09.331Z</atom:updated>
            <content:encoded><![CDATA[<h3>Net AI selected by Digital Catapult to support trusted Telecom AI adoption in national critical infrastructure</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*722gYdv2YjmkDc-GTWCaaw.jpeg" /></figure><p>Net AI has been selected to contribute an Energy Efficiency application to Digital Catapult’s Assured 5G Technical Authority (A5G-TA) programme.</p><p>A5G-TA is part of Innovate UK’s wider work to support the development and adoption of next-generation connectivity technologies across the UK economy. Led by Digital Catapult, the programme will create a vendor-neutral assurance capability for Stand-alone Non-Public Networks, helping organisations understand whether private 5G and AI-native telecoms systems are ready to be deployed in critical infrastructure.</p><p>The project will also create an AI Network Data Catalogue with openly accessible network energy and operational datasets. This will give UK innovators a stronger data foundation for trusted Telecom AI, reducing reliance on closed, external datasets and supporting more sovereign capability in the UK.</p><p>Alongside this, A5G-TA will develop practical methods for assessing whether AI-enabled network operations are trustworthy, explainable, and safe to use without creating new risks for service quality or performance.</p><p>Net AI’s contribution will focus on energy efficiency, one of the clearest and most urgent use cases for AI in telecoms. The company will deploy an Energy Efficiency xApp designed to support more efficient use of radio network resources, while ensuring that any energy-saving decision can be evaluated against its impact on quality of service.</p><p>This is particularly important in radio access networks, where reducing energy consumption is not simply a matter of switching resources off. Traffic demand changes throughout the day, decisions can affect neighbouring cells, and any efficiency gain has to be balanced against service reliability. The xApp will support A5G-TA’s wider work by providing a practical benchmark for assessing AI-enabled energy-efficiency applications, including how much energy is saved, whether service quality is affected, and how explainable the decision-making process is.</p><p>Net AI has previously worked with Digital Catapult through the Technology Access Programme, using the SONIC Labs Open RAN testbed to validate an energy-saving application in a realistic network environment. That experience provides a useful foundation for this new collaboration, which will contribute to A5G-TA’s broader goal of building confidence in trusted, vendor-neutral assurance for AI-native telecoms.</p><p><strong>Alexis Duque, Net AI’s VP of Engineering, said:</strong> <em>Artificial intelligence offers a clear path to reducing telco network energy consumption, but it requires reliability to ensure live service quality remains safe and unaffected. Through the A5G-TA project, we are deploying open testing capabilities and energy-saving solutions within the UK, leveraging years of experience in this field. Net AI is committed to this sovereign approach because it ensures the industry is not dependent on closed datasets, creating a transparent and reliable foundation for trustworthy AI enabling network automation.</em></p><p><strong>Dritan Kaleshi, Director of Technology — Future Networks &amp; Digital Infrastructure, at Digital Catapult said:</strong> <em>“The collaboration with Net AI demonstrates how trusted, vendor-neutral assurance can help turn the potential of AI-native telecoms into practical applications for critical infrastructure. Energy efficiency is an important use case, but organisations need confidence that AI-driven decisions can deliver these benefits without compromising network performance, reliability or service quality. By combining real-world testing with open network data and independent assurance, A5G-TA is helping create the evidence and confidence needed for responsible adoption of these technologies across the UK. This work will also contribute to a stronger sovereign foundation for the development and deployment of trusted Telecom AI.”</em></p><p>By supporting independent, vendor-neutral assurance for private 5G and AI-native telecoms networks, this collaboration will help build confidence in how AI is adopted across critical infrastructure. Net AI is proud to contribute to work that will make high-quality network data more accessible for future UK innovation and strengthen the sovereign capability needed to develop trusted Telecom AI.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=fdc16c29dc12" width="1" height="1" alt=""><hr><p><a href="https://medium.com/net-ai-insights/net-ai-selected-by-digital-catapult-to-support-trusted-telecom-ai-adoption-in-national-critical-fdc16c29dc12">Net AI selected by Digital Catapult to support trusted Telecom AI adoption in national critical…</a> was originally published in <a href="https://medium.com/net-ai-insights">Net AI Insights</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Inside Net AI: A Conversation with Alexis Duque on Turning Research Into Product]]></title>
            <link>https://medium.com/net-ai-insights/inside-net-ai-a-conversation-with-alexis-duque-on-turning-research-into-product-f039e8260338?source=rss-a2d8f7b90894------2</link>
            <guid isPermaLink="false">https://medium.com/p/f039e8260338</guid>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[research]]></category>
            <category><![CDATA[telecommunication]]></category>
            <category><![CDATA[ml-engineering]]></category>
            <category><![CDATA[ai]]></category>
            <dc:creator><![CDATA[Net AI]]></dc:creator>
            <pubDate>Thu, 06 Aug 2026 10:01:03 GMT</pubDate>
            <atom:updated>2026-08-06T10:01:03.933Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*q66csp1S6p9lLgdeKYMKGg.png" /></figure><p>I spoke with Alexis Duque, VP of Engineering at Net AI, about how he joined the company in its earliest stages, what it means to turn research ideas into products, and why AI in telecom needs to be approached carefully. We also discussed PAISES-6G, the limits of full automation, and why sometimes the most useful role for AI is not to replace humans, but to help engineers manage complex networks more effectively.</p><h3>Life in Lyon and Joining Net AI</h3><p><strong>Mariam:</strong> Tell me a bit about where you’re based and what’s the best thing to do there.</p><p><strong>Alexis:</strong> I’m based in Lyon, in France, and my favourite thing to do is to get my gravel bike and go to the forest or the Monts d’Or. There are some small mountains nearby, and it’s beautiful. I think what I like about Lyon is that you are in the city, but you can be in the countryside very quickly.</p><p>What else can I recommend doing in Lyon? It depends. I like the Fine Arts Museum there. I think it’s the second biggest in France after the Louvre, so it’s a big one. Culturally, there’s a lot to do as well, especially around music. Lyon has a great opera house and a really nice auditorium.</p><p><strong>Mariam:</strong> That’s great. Cycling seems to be a bit of a Net AI theme, especially between you, Paul and Rokas. Lyon sounds like a good place for it too. Can you tell me a bit about how you joined Net AI? What was your first introduction to the company?</p><p><strong>Alexis:</strong> Yes. So much time has passed since then. It will be six years in September.</p><p>I was thinking about what I wanted to do with my job. I discussed it with my friend, Razvan, who’s been my PhD advisor, and it appeared that he knew Paul, because he’s Romanian as well, and they were both working in the telecoms field.</p><p>During my PhD years, I always wanted to work with Paul in some way. He was not yet working in AI and was more focused on IoT and cybersecurity, and I was also doing that in my former company. So I was closely following his work at the University of Edinburgh.</p><p>At the time, Paul was looking for someone to start the entrepreneurship process because the company was not created yet. It was just an idea. They had the patent and the paper, and they wanted to make a company based on that. He got a grant from Innovate UK through the ICURe programme, which helped him begin the process of market discovery and setting the company’s direction.</p><p>When Razvan found out about this, he thought of me. I was excited about the idea of creating a company from scratch. I was also looking at AI applied to the Internet of Things, because that was my main focus. And I was excited about working on AI for telecoms as well, working with Paul.</p><p>So I had an interview, and we started working together in September 2020, during the COVID-19 years. Because everything was remote, I didn’t have to move to Edinburgh, which probably made it possible for me to take the opportunity at that stage.</p><p>I wasn’t against the idea of moving to Edinburgh in general, but at the time there wasn’t really a company yet. My contract was only from September to December, and if things went well, it could be extended until around May the following year. So moving country for something that was still so early would have been difficult.</p><p><strong>Mariam:</strong> Were you already done with your PhD when Net AI started?</p><p><strong>Alexis:</strong> Yes. I defended my thesis in 2018, so two years before that.</p><h3>What a VP of Engineering Actually Does</h3><p><strong>Mariam:</strong> What does your role as the VP of Engineering involve?</p><p><strong>Alexis:</strong> I do a lot of things, and I think VP of Engineering is a bit of a shortcut. My main responsibility is to shape the technical team and the engineering process, and to help the team do their best when writing code. So I spend a lot of time reviewing code.</p><p>I give them ideas and unblock them when they’re stuck. Another big part of my role is turning ideas into something industry-ready. I mean something that is engineered, not just a research paper, but software that can actually be packaged and used. So, in a way, I help turn research papers into software.</p><p>I also manage a couple of European projects. I attend meetings with partners, prepare deliverables, join technical discussions, and distribute the work across the team. So I absorb a bit of the administrative side of the European projects: the meetings, the coordination, and so on.</p><p>Also, when I have time, and it’s a bit of a background task, I look at what the industry is doing, what competitors are doing, and what is happening in standards. That includes what people at O-RAN and 3GPP are working on, how it works, and how we can quickly identify which API or interface we need when a customer has a specific requirement.</p><h3>Turning Research Into Software</h3><p><strong>Mariam:</strong> You work closely with Orlando, Sam, Ana and Alec in ML research and ML engineering. I had this question further down the line, but since you pointed it out already, Net AI is very closely connected to academia, and a lot of the team’s work starts from research, ideas, and papers.</p><p>I’d like to talk about how that translates into the business world. You mentioned turning ideas into a technical product. How do you turn research into something operators can use and explain what we do in simpler terms without oversimplifying it?</p><p><strong>Alexis:</strong> Yeah, I think we can see research ideas a bit like a raw diamond. To be able to sell it, we first need to polish it. I would say that’s partly the role of the ML researcher and ML engineer team. Then we need to package it and make something nice, something that shines, a good package to send to the customer.</p><p>That’s not the whole process, but it’s quite exciting and interesting to refine a research idea and make it something that a customer can buy, while also being able to simplify it and explain it without making it too complex, but also not too simple.</p><p><strong>Mariam:</strong> And what is the step-by-step for that? Does an ML researcher, for example, read a paper and say, “I found this cool thing, maybe we can integrate it into our solution”?</p><p><strong>Alexis:</strong> Yes, I would say someone like Orlando, Ana, or Marco will point out a research paper, look at it, and then brainstorm how we can exploit it, modify it, or use it as it is in our project.</p><p>Then Sam would run some experiments and implement the model in our pipelines, train it, and write the code necessary to use it in a product.</p><p>So that’s the second step. And the third step is to make it something we can deploy and sell as a product. That’s more the role of full-stack engineers: building the backend to run the model, building the interface, and creating a broader API for the user to interact with the machine learning model.</p><p><strong>Mariam:</strong> This is just a question that came to my mind, but how does IP work in that case? When research builds on existing papers and ideas, where is the line between learning from the field and creating something new?</p><p><strong>Alexis:</strong> Yes, because most of the time, when I say we read papers, it’s not about taking one paper and turning that directly into a product. We read a lot of papers, and by reading all these papers and all these ideas, they spark other ideas that we want to explore.</p><p>Sometimes we read something that’s not related to telecoms, maybe a paper using AI in a totally different field. By keeping up with what they are doing, and because we know the telecom space, we know how to take inspiration from those ideas, modify them, and make them suit our needs in telecom. That’s a bit the ML researcher part of the job.</p><p><strong>Mariam:</strong> Yeah, that makes sense. I guess in university, that’s also how writing a dissertation works, right? You do a lot of research into other work, but then you create something unique from what you’ve learned.</p><p><strong>Alexis:</strong> Yes, exactly.</p><h3>AI in Telecom: Useful Tool, Not Full Automation</h3><p><strong>Mariam:</strong> Let’s move on to a more general topic of AI and telecom. We’ve talked about this subject extensively before: how AI adoption in telecom is still difficult, and there are a lot of moving pieces, for example, difficulty introducing something new into legacy infrastructure, etc. With this in mind, I wanted to ask you: what is something that people sometimes misunderstand about AI? Where do you think it can actually be useful in telecom?</p><p><strong>Alexis:</strong> I think initially there were too many expectations for AI. At the beginning, people really believed mobile networks could become autonomous, and everybody was expecting a lot. So the deception was huge. And now they want to put it everywhere.</p><p>Of course, it doesn’t work like that. You should go step by step. I don’t believe that right now we are ready to have fully autonomous networks using AI, but I think it can help with very specific tasks.</p><p>I think it’s more like a tool for telecoms now. For example, root cause analysis is a nice use case that is easy to implement in real life and in the short term. A Network Digital Twin is for experimentation, trying different network configurations without making them totally autonomous, but just making experiments in a controlled environment that replicates reality.</p><p>For energy saving, maybe not putting it as a tool to automate the network and optimise the energy gain automatically, but more to fine-tune the network, even with small gains.</p><p>I see it a bit like a coding assistant with AI. We are not asking a coding agent to write the whole code in your application, but you need a tool, step by step, reviewing it and making telecoms more efficient. Not total full automation, but helping people to better manage the network, rather than giving everything to AI and expecting the network to work for free, and that’s it.</p><p><strong>Mariam:</strong> Do you think the problem with full automation is that it can overlook some issues that a human would not overlook?</p><p><strong>Alexis:</strong> There is this risk, and also the fact that I think networks are not ready to be totally autonomous right now. They don’t have everything needed for that. Some things still need to be manually done, so it would be a mess. And data is not always collected in a consistent way. It’s very complicated to have something that works autonomously without someone looking at it.</p><p>There is also a lot of responsibility with operating a mobile network. I remember in France some time ago, the emergency number was down for a couple of hours. It was just a mistake with a cable, a mistake somewhere in a box, and some people were not able to go to the hospital in an emergency because of that mistake. There is an operator responsible for that.</p><p>So that’s also part of what makes it frightening to make something totally autonomous when you have such huge responsibility. Quality of service, financial penalties, and image are all involved.</p><p><strong>Mariam:</strong> Recently Paul made a point on the <a href="https://substack.com/home/post/p-207786231">Sparring Partners podcast</a> that AI is not really needed for everything, and in many cases it can be an overkill. That’s what prompted me to ask you this question, because we do see it everywhere, and we hear a lot of companies use it all the time.</p><p>But again, this could be a case of big hype around it. Everyone spent a lot of money, and now they’re trying to jam it into every type of operation possible to make a return on investment.</p><p><strong>Alexis:</strong> Recently, I had lunch with Razvan, the friend I mentioned before, and he was working on something for a company. It was a type of research project, and they were doing root cause analysis. They wanted to use AI.</p><p>But the thing is, it was just so simple that it didn’t make sense to use AI. A simple static sequence model did the job. And when the static sequence model didn’t do the job, it was just because they didn’t have the data to be able to get the root cause. So even AI couldn’t do that. Sometimes people want AI just because it sounds good.</p><h3>PAISES-6G and the Next Step for AI in Networks</h3><p><strong>Mariam:</strong> Can you talk a bit about the work we’re doing across our current projects? It would be nice to touch on PAISES-6G, for example, because it’s newly launched and we haven’t really written much about it. If you’re involved in that, what are your thoughts? What’s happening?</p><p><strong>Alexis:</strong> Okay, so PAISES just started. We had the kickoff in June, and we had only one meeting two weeks ago. So it’s super new, but I’m super excited about it. It’s a new project with a lot of partners. It’s one of the biggest projects that I have ever participated in.</p><p>It’s about cybersecurity for 6G, and anomaly detection will be our main focus in the project. We are leading a task on that, and most of our efforts will be on a work package where we will develop anomaly detection for autonomous networks and agentic AI for telecoms.</p><p>So it will probably be some agent to flag a cyber attack or anomalies. That can be a cyber attack, or network misconfiguration, or maybe another model or another agent willing to do something super awkward to the network, and we spot this.</p><p><strong>Mariam:</strong> Does it connect to a previous Horizon Europe project, like ORIGAMI for example? Because again, it’s funded by them. ORIGAMI’s main message was to tackle the barriers that we’re going to encounter with 6G. And this focuses on cybersecurity. So I was wondering if they’re connected.</p><p><strong>Alexis:</strong> It’s in the Horizon Europe programme, so it’s like the next step. I think we can, in some way, think that it builds upon some work that has been done in 5G, proposing more AI-compatible 5G and 6G. And on top of that, building a kind of agentic AI into 6G with a focus on cybersecurity.</p><h3>And Finally: A Recommendation</h3><p><strong>Mariam:</strong> That’s all the main questions I had, but I wanted to end with a recommendation. What is a recent book, film, podcast, or piece of media you enjoyed?</p><p><strong>Alexis:</strong> I just read a really nice book that a friend recommended to me. It’s a comic, but a serious one, about Iran. It’s the story of a girl living through the Iranian Revolution in the 80s.</p><p><strong>Mariam:</strong> Are you talking about <em>Persepolis</em>? I love it. It’s one of my favourite graphic novels.</p><p><strong>Alexis:</strong> Yes! And I think the story goes beyond Iran. It brings a universal message about life, and Marjie is someone many people can relate to in some way.</p><p>I also studied the film in high school, so I had some old memories of it. Seeing the story again almost 20 years later gave me a very different understanding. Now I think it’s an even better one than I did back then.</p><p><strong>Mariam:</strong> I agree. It’s really well written, and with any well-written book, you can find yourself in it in some way. Thank you for that. I’m very happy that’s your recommendation.</p><p><strong>Alexis:</strong> So you will make some more publicity for <em>Persepolis</em>.</p><p><strong>Mariam:</strong> Yes! That’s the biggest takeaway I got from this. Thank you, Alexis.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*S-fVsVhMVZDt4iU0HK401Q.jpeg" /><figcaption>Monts d’Or, captured by Alexis</figcaption></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f039e8260338" width="1" height="1" alt=""><hr><p><a href="https://medium.com/net-ai-insights/inside-net-ai-a-conversation-with-alexis-duque-on-turning-research-into-product-f039e8260338">Inside Net AI: A Conversation with Alexis Duque on Turning Research Into Product</a> was originally published in <a href="https://medium.com/net-ai-insights">Net AI Insights</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Reassessing Open RAN]]></title>
            <link>https://medium.com/net-ai-insights/reassessing-open-ran-28d7dacbdb32?source=rss-a2d8f7b90894------2</link>
            <guid isPermaLink="false">https://medium.com/p/28d7dacbdb32</guid>
            <category><![CDATA[telecommunication]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[oran]]></category>
            <category><![CDATA[ai]]></category>
            <dc:creator><![CDATA[Net AI]]></dc:creator>
            <pubDate>Thu, 12 Mar 2026 23:17:26 GMT</pubDate>
            <atom:updated>2026-03-12T23:17:26.214Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*bucELG7w-_IJ97gogvpMOQ.jpeg" /></figure><p>Over the past year, we’ve seen the Open RAN conversation move in different directions.</p><p>On one hand, industry forums remain optimistic and paint a positive image of O-RAN, claiming it’s on its way and the MNOs are eagerly subscribing to the ideology. During a recent <a href="https://lr-resources.lightreading.com/free/w_defa10122/?p=w_defa10122">Light Reading webinar</a> discussing the results of the Winter 2025 Heavy Reading Open RAN Operator Survey, moderator Gabriel Brown (Omdia) led a discussion on architecture and ecosystem maturity. The survey results highlighted progress in standardisation, expanding lab environments, and increasing operator engagement. According to the findings, 60% of operators identified Open RAN as their primary architectural choice for 6G, and many are either adopting O-RAN principles today or actively considering their inclusion in future roadmaps.</p><p>Yet the tone on the ground can feel different. Walking the halls of Mobile World Congress earlier this month, Open RAN was far less prominent (if not missing altogether) in conversations than it had been in previous years. The industry’s attention appeared to shift toward AI, automation, and network monetisation, with Open RAN mentioned more as an architectural component than as the headline topic it once was.</p><p>That shift is also reflected in recent industry headlines, which often focus on retrenchment. Vendors once positioned at the forefront of Open RAN have stepped back from hardware ambitions. <a href="https://www.lightreading.com/5g/nec-retreat-from-5g-looks-like-another-big-blow-to-open-ran?utm_source=Sailthru&amp;utm_medium=email&amp;utm_campaign=Issue:%202026-01-13%20Light%20Reading:%205G%20%5Bissue:80658%5D&amp;utm_term=Light%20Reading:%205G">NEC announced plans to discontinue its 4G and 5G radio equipment business</a> and refocus on software. Mavenir, long positioned as a champion of Open RAN hardware, <a href="https://www.lightreading.com/open-ran/mavenir-boss-regrets-open-ran-bet-in-u-turn-after-financial-rescue">exited the radio market</a> following financial restructuring. Other players that once expressed ambitions to act as full-scale Open RAN system integrators have since narrowed their focus, choosing to specialise in specific parts of the value chain rather than deliver end-to-end RAN solutions. After these developments, I couldn’t help but wonder, is there still hope for O-RAN long-term, or is it quietly receding?</p><h3>Where it started</h3><p>Before exploring the challenges that followed, let’s quickly recap how and why O-RAN emerged in the first place.</p><p>By the mid-2010s, telecom operators were getting frustrated with tightly bundled network stacks that were expensive to run, slow to evolve, and difficult to manage, often tying them to a small number of suppliers. At the same time, the IT world was moving toward running software on general-purpose servers, using virtualisation to decouple applications from specific hardware. Operators began applying similar thinking to telecom networks, starting in the core and gradually pushing those software-based, disaggregated approaches closer to the radio network. In 2018, the O-RAN Alliance was formed to define open, standardised interfaces that promised to enable multi-vendor RAN deployments and increase competition.</p><h3>Where we RAN into some issues</h3><p>So why did O-RAN fail to instantly disrupt the market, as it promised? There are a few reasons:</p><ul><li><strong>Operational complexity in the critical infrastructure.</strong></li><li><strong>Multi-vendor system integration.</strong></li><li><strong>Policy constraints and uneven government support.</strong></li></ul><p>Theoretically, the RAN Intelligent Controller (RIC) is meant to make network management easier. An operator should be able to deploy software applications to adjust network behaviour without modifying the underlying hardware. This sounds quite straightforward. In practice, however, the telecom reality is more complex.</p><p>On an iPhone, the iOS operating system provides standard interfaces that allow applications from different vendors (Google, Slack, Instagram, etc.) to run side by side. If one app underperforms, it can be deleted and replaced with another in seconds. The RIC is conceptually similar: it provides a standardised control layer where different xApps and rApps can be deployed to manage networks and optimise their performance. However, telecom networks are critical infrastructure. Replacing or modifying a software module is not as simple as uninstalling an app; it carries operational risk, regulatory considerations, and the potential for service disruption. Even with standardised interfaces, plugging components in and out of a live RAN environment requires careful validation. Tools such as digital twins may help by allowing operators to test applications in simulated environments before deployment. <em>If</em> the RAN becomes fully software-defined, operators could eventually introduce or replace applications during low-traffic periods with minimal disruption. Telecom networks, however, are not run like consumer IT platforms, and the transition toward that level of flexibility is gradual. The principle of software-driven openness has been compelling for over a decade, but its implementation is constrained by the operational realities and cost structures of carrier-grade networks.</p><p>While O-RAN defines standardised interfaces, openness does not automatically translate into plug-and-play interoperability. Unlike consumer software platforms with mature APIs and tightly governed ecosystems, parts of the O-RAN stack rely on evolving specifications and often RIC vendor-specific software development kits (SDKs). Standard compliance does not guarantee uniform implementation maturity, and operators often face additional validation and integration work before deploying multi-vendor components at scale. This gap between architectural intent and implementation maturity helps explain why operators continue to cite system integration as a primary challenge.</p><p>According to a <a href="https://s3.us-east-1.amazonaws.com/publicmarketingprograms.qtxasset.com/Fierce_Network_Open+RAN+Research+Report_Final.pdf">Fierce Network Research survey,</a> 37% of respondents identified system integration as the most challenging aspect of the Open RAN architecture to implement. While disaggregation expanded vendor choice, it also introduced additional coordination across interfaces and components.</p><p>The experience of large technology suppliers illustrates this challenge. A few years ago, Dell Technologies announced ambitions to act as a system integrator for Open RAN, positioning itself as the party that would assemble, deploy, and operate multi-vendor RAN solutions. At the time, its leadership described the industry as swinging from one extreme to another: first, tightly integrated systems with limited choice, and then Open RAN, where choice increased but so did complexity. The expectation was that a neutral systems integrator could help strike a balance, but in practice, this proved difficult to sustain. Dell eventually stepped back from this role and returned to supplying general-purpose hardware, later partnering with Ericsson to support cloud-based deployments. As one industry observer put it, delivering an end-to-end RAN, covering radios, baseband software, hardware platforms, and integration, is simply too large an undertaking for most companies, making specialisation a more viable strategy.</p><p>Another factor frequently cited as slowing Open RAN adoption is the lack of consistent government support. In the same Fierce Network survey, a majority of respondents said regulation or government backing was important to accelerating Open RAN, with 43% calling it very important and 22.5% describing it as critical. John Baker (founding member of the Open RAN Policy Coalition) echoed this view, arguing that vendor concentration in major markets has increased rather than decreased, raising questions about long-term supply resilience. Despite these concerns, broader geopolitical and economic issues currently rank higher on government agendas, limiting the level of direct intervention or support aimed specifically at expanding competition in the telecom vendor ecosystem.</p><h3>Where Open RAN Stands Today</h3><p>Perhaps unsurprisingly, rather than a sweeping, overnight transformation, Open RAN adoption is shaped by operator risk tolerance and market structure.</p><p>AT&amp;T was once one of the most vocal champions of Open RAN and played a central role in shaping its early momentum. However, the company has opted to anchor its network modernisation around a single large supplier rather than actively pursue a multi-vendor approach. Unlike <a href="https://www.theverge.com/23297756/5g-rakuten-mobile-ceo-oran-cloud-network-decoder">Rakuten</a>, which built Open RAN from scratch in new networks, AT&amp;T is introducing Open RAN gradually within a large, established network. To reduce risk, it has chosen to work closely with a single major supplier as a foundation, using Open RAN–compliant radios and cloud-based network software, while keeping performance and service stability as top priorities. This does not necessarily signal a rejection of O-RAN principles, but it does illustrate a shift to pragmatism. Given the challenges discussed above, that decision is not difficult to understand. Managing a nationwide, live network leaves little room for experimentation. It looks like AT&amp;T and Ericsson are treating Open RAN as an evolution of the existing network, instead of an instant replacement.</p><p>Meanwhile, Taiwan offers a useful example of where Open RAN has gained traction under the right conditions. Government policy has played a central role, with spectrum made available and clear guidance encouraging local companies to focus on enterprise use cases rather than large national networks. This has created momentum around Open RAN in private 5G deployments, where enterprises can benefit from greater flexibility, the ability to combine hardware and software from different vendors, and reduced dependency on proprietary systems. In this environment, Open RAN is less about replacing incumbent suppliers at a national scale and more about enabling tailored, cost-effective solutions for specific business needs. Several Taiwanese vendors have built significant Open RAN portfolios as a result, supported by targeted public funding aimed at advancing Open RAN technologies and strengthening the broader ecosystem.</p><p>These examples illustrate that Open RAN is not progressing uniformly across markets. The variation does not necessarily signal inconsistency, but it reflects different starting points and strategic priorities.</p><h3>The Long Game</h3><p>What may be fading is the early narrative of rapid disruption through hardware diversification. What appears to be strengthening instead is a slower architectural shift: open interfaces, software-driven control, and automation frameworks designed to support flexibility over time.</p><p>The telecom industry has seen similar transitions before. IPv6 was standardised decades ago, yet its global adoption required ecosystem-wide alignment across vendors, operators, and equipment manufacturers. Standards alone do not accelerate deployment; maturity, interoperability, and economic incentives do. Open RAN may follow a comparable trajectory: technically viable, strategically endorsed, but dependent on coordinated execution.</p><p>Another structural factor shaping adoption is the broader 5G business environment. Despite early enthusiasm, a definitive “killer application” for 5G has yet to materialise at scale. Without significant new revenue streams, operators remain focused on cost control and network stability. In that context, large-scale architectural experimentation becomes harder to justify financially.</p><p>Open RAN is not disappearing. Nor is it unfolding as a sudden revolution. It is progressing unevenly, shaped by policy, economics, ecosystem maturity, and operational realities of critical infrastructures.</p><h3>Explore Further</h3><p><a href="https://www.lightreading.com/open-ran/airspan-seeks-to-capitalize-on-5g-exits-of-mavenir-and-nec">https://www.lightreading.com/open-ran/airspan-seeks-to-capitalize-on-5g-exits-of-mavenir-and-nec</a></p><p><a href="https://www.lightreading.com/5g/nec-retreat-from-5g-looks-like-another-big-blow-to-open-ran?utm_source=Sailthru&amp;utm_medium=email&amp;utm_campaign=Issue:%202026-01-13%20Light%20Reading:%205G%20%5Bissue:80658%5D&amp;utm_term=Light%20Reading:%205G">https://www.lightreading.com/5g/nec-retreat-from-5g-looks-like-another-big-blow-to-open-ran?utm_source=Sailthru&amp;utm_medium=email&amp;utm_campaign=Issue: 2026–01–13 Light Reading: 5G [issue:80658]&amp;utm_term=Light Reading: 5G</a></p><p><a href="https://www.youtube.com/watch?v=r25NiDTZ5gU">https://www.youtube.com/watch?v=r25NiDTZ5gU</a></p><p><a href="https://www.fierce-network.com/wireless/open-ran-going-survive-or-not">https://www.fierce-network.com/wireless/open-ran-going-survive-or-not</a></p><p><a href="https://s3.us-east-1.amazonaws.com/publicmarketingprograms.qtxasset.com/Fierce_Network_Open+RAN+Research+Report_Final.pdf">https://s3.us-east-1.amazonaws.com/publicmarketingprograms.qtxasset.com/Fierce_Network_Open+RAN+Research+Report_Final.pdf</a></p><p><a href="https://www.lightreading.com/open-ran/at-t-is-still-kidding-everyone-about-doing-open-ran">https://www.lightreading.com/open-ran/at-t-is-still-kidding-everyone-about-doing-open-ran</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=28d7dacbdb32" width="1" height="1" alt=""><hr><p><a href="https://medium.com/net-ai-insights/reassessing-open-ran-28d7dacbdb32">Reassessing Open RAN</a> was originally published in <a href="https://medium.com/net-ai-insights">Net AI Insights</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Inside Net AI: A Conversation with Orlando E. Martínez-Durive]]></title>
            <link>https://medium.com/net-ai-insights/inside-net-ai-a-conversation-with-orlando-e-mart%C3%ADnez-durive-e8116059cbe2?source=rss-a2d8f7b90894------2</link>
            <guid isPermaLink="false">https://medium.com/p/e8116059cbe2</guid>
            <category><![CDATA[mobile-network]]></category>
            <category><![CDATA[research]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[critical-infrastructure]]></category>
            <dc:creator><![CDATA[Net AI]]></dc:creator>
            <pubDate>Tue, 27 Jan 2026 10:02:03 GMT</pubDate>
            <atom:updated>2026-07-23T09:24:11.335Z</atom:updated>
            <content:encoded><![CDATA[<h3>Inside Net AI: A Conversation with Orlando E. Martínez-Durive on AI, Mobile Networks, and Critical Infrastructure</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*q66csp1S6p9lLgdeKYMKGg.png" /></figure><p>Net AI is proud to have a dedicated team of diverse professionals working from around the world. To start the year off, we thought we’d share a conversation between Mariam Kalandia, our marketing assistant and Orlando E. Martínez-Durive, a Machine Learning Researcher at Net AI and a post-doctoral researcher at IMDEA Networks Institute.</p><p>In this interview, Mariam speaks with Orlando about his work at the intersection of academic research and real-world deployment. Based in Madrid, Orlando shares his perspective on AI for critical infrastructure, the challenges of deploying intelligent systems in mobile networks, and how initiatives like Open RAN, digital twins, and Safeguarded AI, the project Net AI is undertaking backed by the UK Advanced Research and Invention Agency (ARIA), can help bridge the gap between cutting-edge research and reliable, large-scale industry adoption.</p><h3>From Academia to Net AI</h3><p><strong>Mariam:</strong> Thanks for joining me today, Orlando. Before we dive in, can you tell me where you are based currently, and also, how’s the weather?</p><p><strong>Orlando:</strong> So I’m based in Madrid, Spain. The weather is a little bit cold right now, which I like. Most of the time, it’s very nice. It’s chill, but not too cold.</p><p>Something that I really like a lot about the weather in Spain is that even if sometimes it’s hot, it’s still very dry. That’s something I enjoy compared to places where it’s hot and the humidity is also very high.</p><p><strong>Mariam:</strong> Marco (CTO and co-founder of Net AI) is also based in Madrid. Do you see him often?</p><p><strong>Orlando:</strong> Yeah, I see him very often because I’m also working with him as a postdoc in his group. So we have this other space now where I’m more focused on academic research with him.</p><p>We collaborate a lot and have a lot of work that we want to do together. Worst case, I see him twice per week. Other times, I see him three or four times per week.</p><p><strong>Mariam:</strong> And is that how you first came across Net AI?</p><p><strong>Orlando:</strong> Kind of. I don’t remember exactly when, but at some point, the name popped up somewhere. Maybe it was while talking to Marco or through his LinkedIn posts. And because it was quite a catchy name, I wanted to know more about it.</p><p>So I didn’t ask him directly. I believe I went to social media first and then the website to get a sense of what it was. I liked it, and we’ll probably talk about this later, but it’s this idea of taking something theoretical, like an academic paper, and trying to build something based on that.</p><p>Then I found out that Paul (Net AI’s CEO) was also a PhD student at IMDEA, actually the first one. That gave me a sense of connection. He came from the same place where I was doing my PhD, so it developed a sense of familiarity with someone I didn’t actually know yet.</p><p>I also checked out a few of his papers. Then there was an event where people from IMDEA Networks, including alumni, came together and gave talks about what they’re doing now, their experiences, and how they see the future. When he was giving his talk, it aligned a lot with what I want to accomplish and how I look at the future.</p><p>So that was my first real touch with Net AI. After that, I remember thinking, “Hey, I should keep an eye on these guys. This seems interesting.”</p><h3>Life at Net AI: A Startup Mindset</h3><p><strong>Mariam:</strong> What is your current role within Net AI?</p><p><strong>Orlando:</strong> I’m a Machine Learning Researcher. But because we’re a small company with a startup spirit, we need to wear different hats at the same time. I really like what I do as a machine learning researcher, but I also like being involved in data analysis, processing large datasets, collecting data from open sources, and mixing them. I enjoy taking care of different things. At the same time, thinking about how to present our product to customers in the best way possible. That means developing nice interfaces and making sure the demo clearly represents the potential of our technology.</p><p>These tasks sometimes fall outside what a typical machine learning researcher’s role might be. But they’re very important, because you can have amazing technology with great capabilities, but if you don’t convey those capabilities in a simple way to a customer, you’re not going to make it.</p><p><strong>Mariam:</strong> For a non-technical audience, which is myself, to be fair, could you explain the difference between an ML Engineer and an ML Researcher? I interviewed Sam (ML Engineer at Net AI) a couple of months ago, and he gave his perspective from an engineer’s point of view.</p><p><strong>Orlando:</strong> In an ideal scenario, machine learning researchers are people who try to push the boundaries of knowledge. They come up with new ideas on how models and training can be optimized, and how to build smarter and more powerful models.</p><p>At the same time, they assess the complexities behind the models, identify bias, identify issues, and make sure the model is doing what it’s supposed to do and is aligned with the task. That requires a lot of theory, and people with academic backgrounds may relate to it more, though it’s not exclusive to them.</p><p>On the other hand, machine learning engineers focus on how to deploy these models into solutions. They make sure latency is as low as possible, the model is in the right format, and they handle the technical issues that come up when deploying something in production.</p><p>That’s the theoretical distinction. In reality, the lines are blurry. Sam also takes care of tasks outside a typical ML engineer role — he builds models, analyzes research papers, and ensures they’re as advanced as possible. And I sometimes handle engineering tasks as well.</p><p><strong>Mariam:</strong> Is there a particular project or research area you’re especially excited about right now?</p><p><strong>Orlando:</strong> Yes, I have a couple, and they all fall under the big umbrella of AI for critical infrastructure.</p><p>There are many things within critical infrastructure, but I specifically care about networks. Most of the time, we see networks as a way to watch YouTube, scroll Instagram, or message family. But the same infrastructure is used to reach out to emergency services.</p><p>It’s extremely important that this infrastructure is reliable. On the one hand, the network needs to work perfectly. On the other hand, we have AI models, agentic AI approaches, that promise to change how we approach technology, but at the moment, they’re not always fully reliable. People don’t fully trust them in critical scenarios.</p><p>I like to think that we are working toward a future where we can fully deploy AI agents and trust them to be reliable at all times, especially during emergencies.</p><h3>Why Moving Carefully Matters</h3><p><strong>Mariam:</strong> There’s definitely some reluctance toward adopting AI agents into networks, which is understandable since it’s still a new concept. In your opinion, is this reluctance justified, or should we move forward already?</p><p><strong>Orlando:</strong> I think the reluctance is necessary, and you can look at it from different angles.</p><p>One angle is regulation. Network operators are subject to government regulations and standards, especially regarding emergency services. If a network goes down for a long period, people can’t reach emergency services, and operators can face huge fines. When they receive a license, they agree that their network will support emergency communications.</p><p>Another angle is the quality of experience. The mobile communications landscape is extremely competitive. While only a few major operators own infrastructure, many intermediary companies, like virtual operators, resell access at lower prices.</p><p>These intermediaries earn revenue by acting as a middle layer, meaning major operators earn less per customer while covering the infrastructure costs. So operators need to ensure their service quality is as high as possible to avoid losing customers.</p><p>It’s also expensive to train models. If you want the best ones, you often need partnerships, which take time to form. Doing it alone requires massive investment in data centres, training, and data collection. Operators need to be confident they’ll see returns on that investment.</p><p><strong>Mariam:</strong> So basically, they’re being asked to commit before knowing what the results will look like.</p><p><strong>Orlando:</strong> Yes, exactly. You need to commit a lot of resources before having a clear idea of what you’ll get in return.</p><h3>Research, Energy, and Network Intelligence</h3><p><strong>Mariam:</strong> Tell me about your work at IMDEA. You’ve finished your PhD, right?</p><p><strong>Orlando:</strong> Yes, I finished my PhD recently, which makes me very happy. I’m currently working as a postdoctoral researcher at IMDEA Networks. I think I’m in a very good position, combining my role as a machine learning researcher at Net AI with my postdoc.</p><p>As a postdoc, I have time to think about new ideas that leverage mobile networks as large-scale sensors and to improve their operation. As an ML Researcher, I develop and implement state-of-the-art solutions to improve networks while addressing the practical aspects that academic papers often overlook. This last point is something that I always try to keep in mind during my research. Is what I’m doing practical and can be deployed within the current infrastructure, or is it more theoretical and futuristic?</p><p><strong>Mariam:</strong> Can you talk a bit about your research?</p><p><strong>Orlando:</strong> My research focuses on three topics:</p><p>Mobile network metadata analysis. I look at questions like: what services do people use most on their phones? Does that relate to socio-economic variables like income, unemployment, and even political orientation? Is 5G being adopted across all segments of the population, and how fast? What is the impact of major events like protests on mobile networks?</p><p>Routing and network measurements. It surprises me how fragile and resilient the internet is at the same time. On the surface, everything seems to work smoothly, but under the hood, it is a bit chaotic. A single misconfiguration can impact a huge portion of the internet. We see this when cloud services go down, and half the internet is affected.</p><p>Sustainability and energy efficiency. This is something I’m very passionate about, given that RAN deployments are expected to be denser and the need for Edge AI is growing. I have my personal concerns about the future, energy use, and CO₂ emissions.</p><p><strong>Mariam:</strong> How does this research relate to Net AI?</p><p><strong>Orlando:</strong> There’s a lot of academic work showing that mobile networks consume a lot of energy. Operators need to keep antennas running all the time, which increases operational costs.</p><p>My research examines what is currently being done to save energy and what more could be done.</p><p>At Net AI, we believe that there’s a clear gap between what is being done and what’s possible. One specific project we’re working on, funded by ARIA, is building a digital twin to measure and compare energy-saving approaches that build on AI.</p><p>Operators usually test new solutions through small pilots, which are costly and slow. Our digital twin lowers that barrier, allowing AI engines to be tested safely in an environment that closely replicates real networks.</p><p>We’re putting a lot of effort into making the digital twin as realistic as possible. I’m very excited about this because it can be a game-changer. It allows AI models to learn through feedback loops and reinforcement learning, improving with each iteration.</p><p>In this case, we’re building the “game” for the AI to play.</p><p><strong>Mariam:</strong> Yes, it’s part of ARIA’s larger programme called <em>Safeguarded AI</em>, which connects closely to what we discussed earlier. It provides a level of assurance that energy-saving solutions will work and remain safe in large-scale network deployments, lowering the risk for MNOs to adopt them.</p><p><strong>Orlando:</strong> Exactly. Operators can configure the digital twin based on their own networks, embedding their real infrastructure parameters. That gives them confidence that what the AI does in the twin will match real-world deployment.</p><h3>The Future of Intelligent Networks</h3><p><strong>Mariam:</strong> Besides the digital twin project, what else are you working on at Net AI?</p><p><strong>Orlando:</strong> I’m also working on collaborations with major companies. These projects involve making our models compatible with Open RAN and wrapping our capabilities into rApps and xApps.</p><p><strong>Mariam:</strong> From your perspective, where do you see the biggest opportunities for ML research in Open RAN?</p><p><strong>Orlando:</strong> Open RAN de-aggregates network components to avoid vendor lock-in and increase competition. One key component is the Non-Real-Time RIC, where rApps can be deployed without very tight latency constraints.</p><p>This makes it much easier to deploy intelligence in networks. Previously, implementing ML required big changes in the network. Now, the architecture clearly defines where data comes from and how actions are applied.</p><p>The biggest opportunity is this clear framework, which reduces implementation complexity and makes operators more confident when deploying ML solutions.</p><p><strong>Mariam:</strong> Is the network easier to manage under Open RAN?</p><p><strong>Orlando:</strong> The short answer is yes and no. There are advantages, but also challenges. Using components from different vendors introduces security and integration concerns. One weak component can compromise the entire network.</p><p>So while there’s a lot of improvement, it needs to be handled very carefully.</p><p><strong>Mariam:</strong> Some Open RAN vendors have scaled back recently. Does that affect how ML fits into the future of networks?</p><p><strong>Orlando:</strong> I don’t think it changes the big picture. Once people see what’s possible with AI models, they’ll find a way to implement them — Open RAN or otherwise.</p><p>The question isn’t whether it will happen, but how. There’s no way back.</p><h3>And Finally,</h3><p><strong>Mariam:</strong> Okay, my final question is: What’s the most recent book, movie, or piece of media that’s made an impression on you?</p><p><strong>Orlando:</strong> I have two, actually.</p><p>One is a book I read years ago that I have been thinking about a lot recently: <em>The Diamond Age</em> by Neal Stephenson. It’s about a dystopian future where a young girl gains access to a personalized AI device that helps her learn, grow, and navigate life. I read it as a teenager, and it felt like science fiction. Now we’re getting very close to that reality, which still blows my mind.</p><p>The second is a YouTube video called “We’re Not Ready for Superintelligence”, based on the ‘AI 2027’ essay. It discusses the exponential growth of AI and how AI systems might start improving themselves. We’re already seeing early signs of this.</p><p>It raises concerns about safety, bias, and governance. When things move this fast, we need to be extra careful. That video really resonated with me, because things can change very quickly, and it’s important that we understand what we’re building and do it safely.</p><p>Through his work at Net AI and IMDEA, Orlando highlights the importance of building AI systems that are not only powerful but also reliable and ready to be deployed in the real world. His perspective reflects a broader shift in the industry toward responsible innovation, where advanced research, practical implementation, and critical infrastructure needs must align to create lasting impact.</p><h3>Learn more about:</h3><p><a href="https://medium.com/net-ai-insights/developing-safeguarded-ai-with-aria-a23859a56988">Our work with ARIA</a></p><p><a href="https://medium.com/net-ai-insights/from-code-to-culture-ml-ai-adoption-in-telcos-and-staying-connected-in-a-global-team-56577cdf0ef7">Interview with Net AI’s ML Engineer</a></p><p><a href="https://www.goodreads.com/book/show/827.The_Diamond_Age">Book Orlando mentioned</a></p><p><a href="https://www.youtube.com/watch?v=5KVDDfAkRgc&amp;t=35s">Video Orlando mentioned</a></p><p><a href="https://www.linkedin.com/company/net-ai-tech">Follow us on our social media</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e8116059cbe2" width="1" height="1" alt=""><hr><p><a href="https://medium.com/net-ai-insights/inside-net-ai-a-conversation-with-orlando-e-mart%C3%ADnez-durive-e8116059cbe2">Inside Net AI: A Conversation with Orlando E. Martínez-Durive</a> was originally published in <a href="https://medium.com/net-ai-insights">Net AI Insights</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[What Agentic AI Really Means for Telcos]]></title>
            <link>https://medium.com/net-ai-insights/what-agentic-ai-really-means-for-telcos-584a49806174?source=rss-a2d8f7b90894------2</link>
            <guid isPermaLink="false">https://medium.com/p/584a49806174</guid>
            <category><![CDATA[agentic-ai]]></category>
            <category><![CDATA[telecommunication]]></category>
            <category><![CDATA[machine-learning]]></category>
            <dc:creator><![CDATA[Net AI]]></dc:creator>
            <pubDate>Wed, 10 Dec 2025 12:55:41 GMT</pubDate>
            <atom:updated>2025-12-10T12:55:41.986Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*IO62bOetHm4Rpk-2e01evA.jpeg" /></figure><h3>What is Agentic AI?</h3><p>We hear about <em>Agentic AI</em> everywhere these days. Major tech companies are racing to integrate it into their operations, and large consultancies are releasing shiny reports claiming it’s the <em>next big thing!</em></p><p>Over the years, many transformative technologies have reshaped industries. In retail, for example, online shopping was once controversial. Many brands debated whether it was worth the risk. Today, having an e-commerce channel is non-negotiable. The same happened in media: streaming was doubted as uneconomic and niche, and now it’s the standard. Entire supply chains, from content delivery to network capacity planning, revolve around it.</p><p>A similar optimism surrounds Agentic AI. According to <a href="https://newsroom.cisco.com/c/dam/r/newsroom/pdfs/Cisco-CX-Agentic-AI-Research.pdf">Cisco’s report on how Agentic AI will shape customer experience</a>, most surveyed executives believe that vendors adopting AI agents will gain major efficiency and personalisation benefits, while those who delay risk customer churn.</p><p>But what exactly is Agentic AI? What should we expect AI agents to actually do in the context of network operations and, more importantly, what should we <em>not</em> expect them to do?</p><h3>The Buzzword Evolution From “AI” to “Agentic AI”</h3><p>At this year’s <em>Mobile World Congress</em> and <em>Connected Britain</em>, “Agentic AI” or “AI Agents” appeared on banners and slides everywhere. Just a couple of years ago, “Generative AI“ or “Gen AI” dominated the stage. Before that, it was simply “Artificial Intelligence“ or “AI.”</p><p>This rapid rebranding reflects two paradigms:</p><ol><li><strong>Genuine exploration:</strong> as the industry better understands AI capabilities that are seamlessly evolving at a fast pace.</li><li><strong>Marketing refresh:</strong> making old ideas sound new.</li></ol><p>Understanding what an AI agent actually does is essential to evaluating how it might improve operations or deliver a return on investment. It’s easy to see why companies have embraced the term <em>Agentic AI</em>: it suggests a more advanced, purposeful form of intelligence — <em>not just any AI, but a smarter, more autonomous one.</em> The shift in vocabulary gives the impression of technical maturity. Yet, in practice, Agentic AI remains largely an evolution in branding rather than a fundamental change in capability.</p><p>As a matter of fact, there’s no universal definition of <em>Agentic AI.</em> Broadly, it describes multiple AI agents, most often powered by large language models (LLMs), that collaborate via natural language to perform complex tasks. It contrasts previous terminology such as (deterministic) AI — a more general term referring to systems capable of performing tasks that normally require human intelligence like pattern recognition, prediction, or decision-making, and GenAI — a subset of AI focused on creating new content like text, images, code, or designs using patterns learned from large datasets and possibly conditioned on user-defined input or exogenous context variables.</p><p>Agentic AI builds on these same foundations but adds a layer of autonomy and coordination: instead of a single model producing output, <strong>multiple models can plan, communicate, and act together toward defined goals</strong>. Conceptually, this represents a step forward. However, since individual AI models are still not fully trusted or deployed at scale in telecom operations, letting multiple agents “cooperate” only amplifies the existing challenges of reliability, interpretability and control in critical infrastructures such as large-scale, operational telco systems.</p><h3>Telecom Reality — ML Adoption Is Taking More Time Than Expected</h3><p>Despite the buzz, large-scale AI deployments remain rare in the telecom ecosystem.</p><p>Agentic AI may sound like the next big thing, but in practice, its goals overlap with the same automation challenges that telcos have been wrestling with for years. Machine Learning research evolves quickly because it lives in an algorithmic domain, where new code, models, and data can make breakthroughs almost overnight. Telecom players, however, operate in a systems domain: they are highly regulated, risk-averse, and slow to evolve. Integrating any new paradigm into live networks is very expensive and time-consuming. That gap between AI’s pace of innovation and telco’s pace of adoption explains why, at major industry events like <em>Connected Britain</em>, “Agentic AI” dominates the headlines even though real-world deployment remains limited. It’s less about what’s already working and more about showing that the industry is keeping up.</p><p>Again, there is no clear definition for it, but in telecom, Agentic AI could mean diverse agents monitoring network KPIs, diagnosing faults, or triggering reconfigurations independently to then coordinate autonomously. This is an <em>evolution</em>, rather than a revolution. More importantly, it is <em>not</em> an answer to the big problem: while AI is used in production for a few, specific, long-timescale, and non-close-loop tasks such as customer support, analytics, or radio planning, the large-scale deployment of AI in mission-critical network control is not dependable enough to be trusted by the major 5G network operators.</p><p>The actual barriers that need to be addressed are:</p><ul><li><strong>Economic friction:</strong> unclear or delayed ROI — According to <a href="https://www.forbes.com/sites/markcperna/2025/03/24/new-data-41-of-gen-z-workers-are-sabotaging-their-employers-ai-strategy/"><em>Forbes</em></a>, 62% of executives expect at least three years to see ROI from AI investments, and one-third report being disappointed with early results.</li><li><strong>Talent gap:</strong> limited in-house AI expertise and little intersection between the machine learning community and the telecom industry.</li><li><strong>Strategic fatigue:</strong> overpromising and underdelivering pilots.</li></ul><p>This slow pace reflects telecom’s structural complexity. Legacy infrastructure makes experimentation costly and cautious. And Agentic AI is not a solution that directly targets these issues — in fact, it risks having the opposite effect by complicating AI operations and further entangling their integration in network systems.</p><h3>Promise vs Pitfall</h3><p>Letting LLMs coordinate to achieve complex network management tasks also introduces two new major challenges that add to the existing burden of AI operation in telco systems:</p><ol><li><strong>Reasoning vs. mimicry: </strong>LLMs do not truly “reason.” When asked to explain their logic, they often reveal a lack of genuine understanding. They succeed by recognising statistical patterns, not by building causal models. Academic research takes the idea of “agency” far more seriously than most industry messaging suggests. A 2022 DeepMind and University of Oxford paper defines an agent not as something that produces output, but as a system that would behave differently if it understood the world differently. By that definition, most LLM-based “agents” today fall short: they generate responses, but do not adapt goals or models of the world itself.</li><li><strong>Sustainability and scale: </strong>LLMs require substantial GPU resources, energy, and cooling infrastructure — possibly orders of magnitude more than “baseline“ neural network models. Running hundreds of models to control a nationwide network could drastically increase energy costs, at odds with telecoms’ paramount goal of reducing OPEX, energy bills and carbon footprints to increase the current, lean profit margins and align with green regulations.</li></ol><p>Deloitte’s <em>Agentic AI Blueprint for Telcos</em> provides a useful example of how the industry is approaching Agentic AI in practice. Their agent, “Hoopie,” is presented as capable of responding to network incidents and supporting human analysts through coordinated agents. In the demonstration, Hoopie primarily assesses situations and provides recommendations, while decisions remain with human operators, illustrating that autonomy is currently framed as assistive rather than fully autonomous.</p><p>The publicly available material focuses on architectural concepts rather than technical implementation, and does not describe in detail how decision logic is generated or validated. At the same time, Deloitte highlights the significant infrastructure requirements of GenAI, noting that telcos will require “chips, energy, water and money” to compete, which brings attention to the scale and sustainability considerations involved in deploying such systems.</p><p>Similarly, one of the recent TM Forum catalyst projects focused on ‘Agentic &amp; autonomous AI for business excellence’ illustrates how the industry currently envisions autonomy: multiple agents coordinating tasks within a shared framework. In the demo, a “master agent” orchestrates specialised agents such as the “business agent” and the “resource agent,” while an “assurance agent” oversees execution. Although this architecture looks futuristic, the publicly available information offers no insight into <em>how</em> decisions are actually made or validated, leaving the underlying reasoning elusive.</p><h3>Looking Ahead — What’s Actually Worth Watching</h3><p>So, where does AI genuinely create value in telecom today?</p><p>One of the sector’s most pressing issues is shrinking profit margins. The recent Vodafone-Three merger illustrates how operators are striving to expand their customer base while keeping infrastructure costs flat. AI agents may help streamline front-office operations like customer service, but the real efficiency gains will come from machine learning systems that improve network management.</p><p>Tools that predict traffic patterns and allocate network resources proactively, rather than reacting after problems occur, are more effective at reducing over-provisioning, improving energy efficiency, and delivering measurable ROI.</p><p>Telcos’ slower pace of AI adoption isn’t due to a lack of ambition but to the complexity of their systems. Integrating and maintaining any AI in live networks demands specialised expertise that many operators are still building. The real opportunity lies in closing that knowledge gap by focusing on technologies that deliver incremental, data-driven improvements, rather than chasing the next buzzword.</p><p><strong>Sources:</strong></p><p><a href="https://www.ibm.com/solutions/ai-agents?utm_content=SRCWW&amp;p1=Search&amp;p4=296972601110&amp;p5=b&amp;p9=176706261763&amp;gad_source=1&amp;gad_campaignid=22372160349&amp;gbraid=0AAAAA-oKwifJJZKpe_ozI90gJLcr9wunO&amp;gclid=Cj0KCQjwjL3HBhCgARIsAPUg7a7cmOaHkDlVfi_Y6OtqujVvTq35wGNFZw2nd9BWyM_ESoBZTztu9EQaAjs8EALw_wcB&amp;gclsrc=aw.ds">IBM’s AI Agents</a></p><p><a href="https://newsroom.cisco.com/c/dam/r/newsroom/pdfs/Cisco-CX-Agentic-AI-Research.pdf">Cisco Report</a></p><p><a href="https://www.ibm.com/downloads/documents/us-en/1443d5f5af4f532a">https://www.ibm.com/downloads/documents/us-en/1443d5f5af4f532a</a></p><p><a href="https://www.forbes.com/sites/markcperna/2025/03/24/new-data-41-of-gen-z-workers-are-sabotaging-their-employers-ai-strategy/">https://www.forbes.com/sites/markcperna/2025/03/24/new-data-41-of-gen-z-workers-are-sabotaging-their-employers-ai-strategy/</a></p><p><a href="https://inform.tmforum.org/research-and-analysis/reports/new-generation-intelligent-operations-agentic-ai-driven-transformation?utm_source=mkto&amp;utm_medium=email&amp;utm_campaign=AN_EM_0810_Q2">https://inform.tmforum.org/research-and-analysis/reports/new-generation-intelligent-operations-agentic-ai-driven-transformation?utm_source=mkto&amp;utm_medium=email&amp;utm_campaign=AN_EM_0810_Q2</a> Round up&amp;mkt_tok=MDIxLVdMRC04MTUAAAGdWVXwLzGeZC_MJLLye6q7wQIHQj-8LDdSG2s-AEmCPjqk4E8R07HXDVSG7R_8MquOySx3jSqEy95qw3KrwEi83DV1AyskMxQSr5TZ3ssAqLcbZuQ</p><p><a href="https://arxiv.org/abs/2208.08345">Kenton et al (2022), <em>Discovering Agents</em></a></p><p><a href="https://www.deloitte.com/global/en/what-we-do/capabilities/gen-ai-services.html">Deloitte’s Hoopie</a></p><p><a href="https://www.tmforum.org/catalysts/projects/C25.0.816/agentic-autonomous-ai-for-business-excellence">TM Forum Agentic and autonomous AI for business excellence</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=584a49806174" width="1" height="1" alt=""><hr><p><a href="https://medium.com/net-ai-insights/what-agentic-ai-really-means-for-telcos-584a49806174">What Agentic AI Really Means for Telcos</a> was originally published in <a href="https://medium.com/net-ai-insights">Net AI Insights</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Inside Net AI: A Conversation with Samuel Knight]]></title>
            <link>https://medium.com/net-ai-insights/from-code-to-culture-ml-ai-adoption-in-telcos-and-staying-connected-in-a-global-team-56577cdf0ef7?source=rss-a2d8f7b90894------2</link>
            <guid isPermaLink="false">https://medium.com/p/56577cdf0ef7</guid>
            <category><![CDATA[telecommunication]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[team-collaboration]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[energy-efficiency]]></category>
            <dc:creator><![CDATA[Net AI]]></dc:creator>
            <pubDate>Wed, 23 Jul 2025 13:09:26 GMT</pubDate>
            <atom:updated>2026-08-19T11:00:20.154Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*q66csp1S6p9lLgdeKYMKGg.png" /></figure><h3>A Conversation with Samuel Knight</h3><p>In today’s fast-evolving telecom world, engineers like Samuel Knight are not just building smarter models, they’re shaping the infrastructure of tomorrow. In this conversation, I sat down with Sam, a Machine Learning Engineer at Net AI, to talk about what a typical day looks like in our small but ambitious team, the promise and friction of applying AI in telecom, and what keeps him motivated, both professionally and personally. Whether you’re in academia or in the telco industry, this conversation offers insights into where telecom is headed, how machine learning is evolving to meet its needs, and why culture still matters in tech.</p><h3>Life as a remote ML Engineer</h3><p><strong>Mariam:</strong> Hello, Sam. Thank you for joining me today.</p><p><strong>Sam:</strong> Thanks, Mariam. Happy to be here!</p><p><strong>Mariam:</strong> Let’s start with the basics: where are you based currently, and what does a typical workday as an ML Engineer look like?</p><p><strong>Sam:</strong> I’m based in Davis, California, a small university town near Sacramento, the state capital. It’s about two hours away from the Bay Area, which everyone knows about, but people don’t seem to know that much about Sacramento. It’s super nice, quiet.</p><p>Usually, my day starts at around six in the morning. I try to get some overlap with people finishing their workdays in the UK in case there are any blockers or things I need to address. I make coffee at seven normally, eat breakfast. Although timing my breakfast significantly depends on the imminent deadlines. So sometimes, I end up putting that off until I’m really starving, and then I have to rush out of my desk to scarf down some cereal.</p><p>My work as a machine learning engineer depends on the stage of the different projects we work on.</p><p>If we’re in more of an exploratory phase, then it’s usually reading papers and, working on some implementations of those papers, making sure that our models are up to date with the state of the art, or there are individual components from those papers that we can incorporate into our model without having to extract the whole thing.</p><p>It’s also doing code reviews. We have a pretty small team that works on the ML side, and we’re all pretty in tune with what the other is working on. It’s a high-contact environment, which is good.</p><p>If we’re in a more delivery phase, then most of my day can be taken up by looking at evaluation metrics from model training runs that we’ve done, making sure that we’re not seeing any big regressions in terms of performance.</p><p>And if we are seeing regressions, then trying to work out exactly what is going wrong during training. If it’s a data issue, for example, and we need to do some more feature selection or clean some features, etc.</p><p>Sometimes I work on more visualisation stuff, front-end components, and sometimes more back-end stuff as well.</p><p><strong>Mariam:</strong> You said your work day depends on a project phase we’re in. Is there a phase you’d rather be working on, whether it’s exploratory or development?</p><p><strong>Sam:</strong> I think the most exciting times are when we’re delivering something and we get to work as a whole team together collaboratively.</p><p>I think some of the research phases can be more siloed in the sense that you spend time researching, reading a specific paper, looking to implement it, and then going to review, whereas when we’re actually on the road to delivering a solution, it’s usually very collaborative, which is probably the most enjoyable time.</p><h3>Inside Our Projects</h3><p><strong>Mariam:</strong> Are you working on multiple projects right now, or are you fully committed to one?</p><p><strong>Sam:</strong> Right now, I’m focused on the <a href="https://medium.com/net-ai-insights/developing-safeguarded-ai-with-aria-a23859a56988">ARIA project</a>. I’ll leave the detailed explanation to Marco (<a href="https://www.linkedin.com/in/marcofioreita/"><em>Marco Fiore</em></a><em>, our co-founder and CTO</em>), but this project involves building generative models that simulate how network conditions change over time across an entire network.</p><p>Instead of predicting a single future value, like what throughput will be next week, our models aim to generate realistic sequences of data that closely resemble the actual patterns we see. In essence, we’re creating a simulator that can produce believable network data.</p><p>This is really valuable because accessing real network data is often difficult. It’s sensitive and expensive for companies to share or collect. So, if we can generate realistic data ourselves, we can use it to train and test models for other tasks like forecasting or anomaly detection. It helps us scale our work much more efficiently.</p><p><strong>Mariam:</strong> So now you’re in the developmental phase with ARIA. Is it mostly collaborative or still solitary?</p><p><strong>Sam:</strong> Yes, we’ve moved on a little bit from the more siloed, paper implementation phase, and now we’re all working on evaluating what parts of the model work, and what don’t. We’re looking at the evaluation metrics and trying to come up with a final stable model that we can then take on to the downstream tasks together. So that’s been really fun!</p><p><strong>Mariam:</strong> Was there a type of project you were really passionate about in particular, not just right now, but in general during your time at Net AI?</p><p><strong>Sam:</strong> Yes, I was really excited, and still am very excited about our EnergAIze solution and the forecasted models that underpin that.</p><p>We see this theme across the world and industries, of increasing need for energy efficiency, cutting costs, and running things smarter. Although I get the sense that the networking space is a little bit behind in that regard. And it makes sense — network operators are focused on delivering the best possible customer experience, and sometimes that feels at odds with energy savings.</p><p>Obviously, if you start tuning down power transformers and cell towers and then reducing the throughput they can accommodate, for example, then a customer might have their data needs not met, and that would be detrimental, and they might leave the network if they feel frustrated.</p><p>But the thing is, network usage patterns are super predictable. We see clear spikes and drops throughout the day and week. So there’s a real opportunity here. It feels like a missed one, honestly, if we keep running everything at full power <em>just in case</em>.</p><p><strong>Mariam:</strong> You’re basically saying telcos shouldn’t be so worried about user experience, because there’s a way for them to both save energy and not sacrifice and maybe even improve the quality of experience?</p><p><strong>Sam:</strong> The key thing is making sure that there’s enough capacity on the network for the users.</p><p>There are these clear periodic patterns where you get spikes in user demand for the network capacity and decreases in that demand, whether that’s day and night patterns or weekend versus weekdays.</p><p>But the “always on” approach of the past, or at least I think it will be of the past going into the next few years, optimises for making sure that those customers have the capacity needs met.</p><p>And industry has had a sort of a reactive approach in terms of meeting additional capacity of building all this extra infrastructure as demand increases. But, it’s still “always on” as the default.</p><p>Having forecasting models that are specifically optimised to stay above user demand while decreasing energy usage as much as possible, I think that’s the way of the future. I mean, not only do you save energy, but you cut costs, because you use less electricity. So, it makes sense from a financial perspective, as well.</p><p>And I think as these models get more and more sophisticated, it’s only going to make the case of achieving this balance better and better instead of putting all your eggs in one basket.</p><h3>Why Telecom Is Slow to Trust AI and Why That’s Changing</h3><p><strong>Mariam:</strong> Is the fear of decreased quality of experience the only blocker for adopting a solution like EnergAIze, or are there other concerns telcos might have?</p><p><strong>Sam:</strong> Good question, and just to preface this, as an engineer, I don’t get that much interaction with the actual customers. So everything that I’m saying is filtered down through Paul and Marco and things I hear.</p><p>I’d say one of the biggest barriers is simply that it’s a new kind of system. AI and ML aren’t new in general. Industries like banking have used them for years. Even in something as old-school as ATMs, there’s machine learning doing character recognition when you deposit a check. That’s been around for the last twenty years or so.</p><p>So, in some cases, it’s very embedded. In the networking space, I feel there’s been less adoption in general, less trust in these systems to deliver. It’s a huge industry and sort of slow-moving by nature, and there is a hesitancy to adopt these solutions from smaller companies like us, just from the trust factor and the scale.</p><p>I think it’s a matter of individual telcos being willing to trial some of these solutions on a small scale, maybe in a single city, or even a couple of antennas around the city, where you won’t have a huge customer impact as you’re tuning the system to the parameters that you want, and then rolling it out and building it out from there.</p><p>So yeah, trust is probably the biggest factor. But there’s still movement. We’re seeing some telcos start to explore this, even building their own internal pilots. I hope that momentum keeps growing.</p><p><strong>Mariam:</strong> It does make sense. These are large organisations, and adopting something new often means stepping outside familiar territory.</p><p><strong>Sam:</strong> Yes, and another aspect might be that we just recently entered this new era of 5G rollout, and already people are talking about 6G.</p><p>One of the newer capabilities with 5G is that you can start to manage individual cell towers via software. Tune them dynamically. That opens the door to applying ML at a much finer level. But that whole software-defined network management approach is still pretty new.</p><p>So in a way, it’s sort of like a double hurdle to overcome in legacy thinking. Not just applying ML models, but actually doing the custom management of the network, via software, in this fine-grained way. I think that’s another thing that the industry is still moving towards being comfortable doing.</p><p><strong>Mariam:</strong> It’s kind of wild to think that 5G isn’t even fully rolled out, and yet 6G is already in development.</p><p><strong>Sam:</strong> I think it’s one of these cases where these standards, for 5G, 6G… It’s this mashup of all these competing interests, all these massive mega networking corporations that are all trying to lay out how they want the standard to evolve based on the challenges they face and so on. It’s just a constantly changing landscape. So, it makes sense that as one standard crystallises, like 5G, and starts to be fully implemented, all these different stakeholders are thinking towards the future and how they want the standard to change and what sort of things they want out of it.</p><p>In one sense, it’s like a very slow-moving space with how big the corporations are. But in another, there’s constant activity behind the scenes.</p><p><strong>Mariam:</strong> AI’s become such a buzzword lately, especially in telecom. It promises a lot and sometimes it delivers, but now we’re seeing governments stepping in with rules and ethical guidelines. Do you think that’s reasonable, or is it veering into moral panic?</p><p><strong>Sam:</strong> I think there has been this huge intersection between the big advancements in generative AI and public concern. I get it. Trusting the information we see online is already tricky and when AI can generate fake photos, videos, or voices that feel real, that trust erodes even more.</p><p>So I understand the caution from that side.</p><p>But zooming in on what we do. Our models are non-LLM, non-text, very task-specific. They generate realistic network KPIs, or help forecast traffic. How much harm can someone do with that box versus a box that can make someone believe that you said something you didn’t?</p><p>So we’re not caught up in the panic. There’s not much moral ambiguity in simulating throughput spikes.</p><p>Where I do see LLMs showing up in telecom is in customer support tools or internal assistants to help manage network operations. And in that space, I think it’s actually empowering. It’s less about harm and more about productivity.</p><h3>On Net AI’s team culture and working from abroad</h3><p><strong>Mariam:</strong> I want to talk a bit about our team and the way you interact with them while being so far away. How do you manage to stay connected?</p><p><strong>Sam:</strong> I think it’s as simple as having regular meetings, touching base with both people in the ML team and the wider team in our regular weekly meetings.</p><p>I’ve also tried to push for more informal meetups, even if they aren’t work-focused. Since we’re remote-first unless you’re in Edinburgh, that human connection is something you have to consciously maintain. Talking regularly, even just casually, goes a long way.</p><p><strong>Mariam:</strong> Since Edinburgh’s sort of our hub, do you feel your relationship with the team varies based on whether people are there or remote?</p><p><strong>Sam:</strong> Not really. I remember when I was in Edinburgh and we had the old office, the work environment was a lot different. It just is when you’re in person working with other people.</p><p>There’s this very cool environment that you just sort of take for granted. But since I was pretty close with the Edinburgh people, when I moved away, I noticed that difference in the environment, but in terms of being close with the team, that hasn’t changed at all. So I’ve enjoyed that.</p><p><strong>Mariam:</strong> Do you have any rituals or habits that help you focus while working from home?</p><p><strong>Sam:</strong> I try to use timers — usually 30 to 45 minutes — to really lock in and work without distractions. Then I’ll take a short break: stretch, grab a coffee, go for a quick walk. I don’t do it as consistently as I’d like, but it helps.</p><p>And honestly, I start my day with a bit of mindless browsing, news, memes just letting my brain idle before switching into work mode. Everyone has their own system, but I think some kind of routine is key when you’re remote.</p><p><strong>Mariam:</strong> I can’t believe I haven’t asked this yet, but how did you end up at Net AI?</p><p><strong>Sam:</strong> I met Paul at university. I took his IoT course, and that was my fourth year. Then, in my fifth year, I demonstrated for that course. So I helped out with some of the tutorials. We stayed in touch, and after uni I spent a year in a full-stack software engineering role, working in advertising, not really my passion, but the team was great, so it was a great experience, but I was looking for a change.</p><p>Then Paul reached out about a company he was spinning out from his research, looking for engineers. I was immediately interested. I had a solid background in ML from university, so it wasn’t a complete leap, but it was still a big shift day to day. I interviewed and joined late 2021 or early 2022, I think? Feels like ages ago now.</p><p><strong>Mariam:</strong> What’s it like having your former professor as your boss?</p><p><strong>Sam:</strong> Surprisingly chill! He was already an authority figure, so the dynamic was familiar. Honestly, Paul’s easy to work with in both roles, and I think the whole team would say the same.</p><p>And the team that we’ve built here, I think, is of like-minded people who are super collaborative, friendly. I hope we can scale that up as we grow, but that’s one of the things I think we should hold on to. The type of people that we have here. It’s a good place to work.</p><h3>The Personal Side</h3><p><strong>Mariam:</strong> What is something people would be surprised to learn about you?</p><p><strong>Sam:</strong> Oh, that’s an interesting question. Something surprising to hear about me…</p><p>I don’t like saying it because I feel like it’s sort of cliché of anyone who’s done this will tell everyone they’ve done it, but I’ve been skydiving before, which was something I wasn’t comfortable doing. But I did it anyway. It’s totally not my thing. I’m not into adrenaline, but my girlfriend was super excited about it and convinced me. I was quietly enthusiastic too, but definitely terrified. One time was enough!</p><p><strong>Mariam:</strong> That’s pretty cool! And it’s definitely on my list too. I was also surprised to find out you lived in Switzerland.</p><p><strong>Sam:</strong> Oh, that is a good answer. I lived in Switzerland; I was born there. My mom’s from California, my dad’s from Scotland. They met while working in Switzerland, and my siblings and I all grew up there. I still go back often. My parents live there, so it holds a special place in my heart.</p><h3>Media Recs &amp; Nerd Outs</h3><p><strong>Mariam:</strong> Any books, podcasts, or shows you’ve been enjoying lately?</p><p><strong>Sam:</strong> From the shows, even though I’m sort of late on the bandwagon, I started watching Succession, which is so much fun. It’s just very entertaining watching all these hyper wealthy people tear their hearts out. I think we just finished season one, so we’re making our way through slowly.</p><p>As for books, after watching <em>Dune Part Two</em>, I picked up the series again. I’d read the first one years ago, but now I’m working through Children of Dune. I’d forgotten how much I love sci-fi — it’s way easier for me to finish fiction than non-fiction. Research papers are another story… not quite as engaging!</p><p>I also discovered George Saunders recently. His short stories are incredible. I read three of his collections back-to-back and would recommend anything with his name on it.</p><p><strong>Mariam:</strong> Those are some great recommendations. Thanks, Sam!</p><p><strong>Sam:</strong> Thanks, Mariam. This was fun!</p><p><strong>Connect with Us</strong></p><p><a href="https://netai.tech/">Net AI</a> is a network intelligence company building advanced machine learning solutions to help telecom operators run smarter, more efficient networks.</p><p><a href="https://www.linkedin.com/in/samuel-knight-11236b139/">Samuel Knight</a> is the Machine Learning Engineer at Net AI. He designs and trains models that are the cornerstone of our network optimisation solutions.</p><p><a href="http://www.linkedin.com/in/mariam-kalandia-ab6573221">Mariam Kalandia</a> is a Marketing Assistant at Net AI. She manages the social media page and led this interview.</p><p><strong>Explore More</strong></p><p><a href="https://mailchi.mp/netai/20d1ndkil4">EnergAIze solution Sam mentioned</a>.</p><p><a href="https://meetings-eu1.hubspot.com/paul-patras?uuid=9ea82cf4-baa9-40f6-97f0-4bec70b613b1">Book a meeting with the CEO and founder of Net AI, Paul Patras, to find out more about what we do</a>.</p><p><a href="https://netai.tech/news">Stay up to date on what we’re up to.</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=56577cdf0ef7" width="1" height="1" alt=""><hr><p><a href="https://medium.com/net-ai-insights/from-code-to-culture-ml-ai-adoption-in-telcos-and-staying-connected-in-a-global-team-56577cdf0ef7">Inside Net AI: A Conversation with Samuel Knight</a> was originally published in <a href="https://medium.com/net-ai-insights">Net AI Insights</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Unlocking Accurate QoE Predictions in Modern Networks]]></title>
            <link>https://medium.com/net-ai-insights/unlocking-accurate-qoe-predictions-in-modern-networks-9d5975546a7c?source=rss-a2d8f7b90894------2</link>
            <guid isPermaLink="false">https://medium.com/p/9d5975546a7c</guid>
            <category><![CDATA[network-quality]]></category>
            <category><![CDATA[service-quality-assurance]]></category>
            <category><![CDATA[qoe]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[machine-learning]]></category>
            <dc:creator><![CDATA[Net AI]]></dc:creator>
            <pubDate>Thu, 10 Jul 2025 13:47:59 GMT</pubDate>
            <atom:updated>2025-07-10T13:47:59.735Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*47xFZ1xD17QlGaKJ8dy2uA.png" /></figure><p><em>How well do operators understand quality of experience?</em></p><p>Despite the telecom industry’s focus on coverage, capacity, and throughput, many operators still struggle to answer a simple but crucial question: <em>how good is the experience for the user?</em> With service provider switching easier than ever, network quality is a key metric that can serve as a proxy for customer retention, revenue and reputation.</p><p>To improve users’ quality of experience, telcos are seeking to develop strong QoE assessment strategies based on various tools. Yet, most current approaches rely on fragmented KPIs, outdated per-network element models, or intrusive crowdsourcing tools that are losing viability due to tightening privacy standards. These methods offer a partial, often delayed view of the customer experience and rarely allow operators to act <em>before</em> problems escalate.</p><p>With the current state of QoE forecasting in mind, we developed DeepQoE, a neural engine designed to predict service quality across a full deployment, enabling operators to understand and proactively improve quality of experience at scale.</p><p>In this post, we explore why QoE matters now more than ever, dive deep into current solutions’ shortcomings and explain how a predictive, metadata-driven approach can give operators the edge in today’s highly competitive landscape.</p><h3>The Challenge: Why Service Quality Remains Poorly Understood</h3><p>Imagine you have a long train journey ahead of you and you want to relax and watch the latest season of “The Bear” (as relaxing as the chaos of a dysfunctional kitchen can be — but hey, each to their own). You didn’t download the episodes in advance, the train’s Wi-Fi is unusable, and despite your phone showing four signal bars, streaming is impossible without buffering stalls.</p><p>When we talk about network quality, we’re not just talking about signal bars or download speeds. Quality of Experience is about how well real-world services perform for end users on a continuous basis. It’s how smoothly a video streams, how quickly a website loads, how responsive a game feels, or whether a video call drops mid-sentence. With 5G now widely deployed and 6G on the horizon, the bar for performance is higher than ever.</p><p>For operators, understanding QoE is critical for:</p><ul><li><strong>Customer retention</strong>: As switching between providers has become frictionless, customers won’t tolerate poor or inconsistent experiences. They leave.</li><li><strong>SLA enforcement</strong>: For enterprise and wholesale customers, delivering on quality commitments depends on having measurable, provable QoE insights.</li><li><strong>Operational planning</strong>: From traffic shaping to cell upgrades, network investments are only effective when guided by accurate, forward-looking quality data.</li></ul><p>And yet, understanding service quality at scale is incredibly complex. Modern networks consist of thousands of sites, with users constantly moving between them. Congestion can spike unpredictably. Devices vary widely in capability. KPIs flood in from every direction, but don’t tell the full story on their own.</p><p>What operators need is a way to see the full picture, predict problems <em>before</em> they impact the user, and make smarter decisions to optimize quality end-to-end.</p><h3>Current Industry Approaches and Their Shortcomings</h3><p>So, what are telcos doing to make sure we can enjoy our favourite shows on trains? Unfortunately, not quite enough. At least, not yet.</p><p>Despite the growing urgency, most QoE solutions in use today fall short, often due to being reactive, fragmented, or unsuited to modern privacy and deployment realities. Here’s a look at the main categories:</p><ol><li><strong>Off-the-shelf forecasting models</strong></li></ol><p>One common approach is the use of off-the-shelf forecasting models, typically deployed on a per-element basis (e.g., per site or per cell). These models are usually trained independently and often rely on historical traffic statistics. The result is a disjointed view of the network, with limited coordination across elements. Because they fail to capture broader patterns or network-wide dynamics, these models frequently underestimate demand, which leads to missed congestion events and ultimately degraded quality of experience. More importantly, they offer no unified visibility into overall user experience, making it difficult for operators to act strategically or proactively.</p><p><strong>2. Crowdsourced data platforms</strong></p><p>Another widely used method is crowdsourced analytics, with platforms like Umlaut aiming to offer a user-centric perspective on QoE. While appealing in theory, this approach depends on users installing monitoring apps that passively collect network data. In reality, this results in incomplete and noisy datasets, which are hard to trust for critical decision-making. This is further curtailed by escalating data privacy provisions, as the new iOS and Android policies limit the scope of background data collection. This makes crowdsourcing not only unreliable but increasingly unsustainable in the long run.</p><p><strong>3. Manual rule-based systems</strong></p><p>Lastly, many operators still rely on manual, rule-based systems to monitor performance, using static thresholds (e.g., “trigger alarm if latency exceeds 100ms”) to flag problems. These systems are rigid and struggle to adapt to real-world variability in user behavior, traffic patterns, and device performance. Worst of all, they are inherently reactive, only flagging issues after service degradation has already occurred.</p><p>Each of these approaches has contributed something to the evolution of QoE monitoring, but none of them provide what operators now need most: accurate, scalable, and predictive insights into real service quality.</p><h3>Introducing DeepQoE</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*XNtE8v9mY6oLhWwADL1hMg.png" /></figure><p>Our approach to accurate quality of experience forecasting? <strong>DeepQoE</strong> — an AI-powered forecasting engine designed to give operators a clear, predictive view of service quality across their networks.</p><p>Rather than relying on fragmented or crowdsourced data, DeepQoE analyses multiple KPIs primarily at the network level, with the option to incorporate user-level metrics to generate a simple yet powerful Quality Index, ranging from 1 (poor) to 5 (excellent). This index can be calculated per user, per cell, or across an entire deployment, enabling operators to visualise quality trends, detect emerging issues, and act preemptively. Whether it’s identifying congested cells, planning capacity upgrades, or triggering intelligent handovers, DeepQoE provides the forecasting intelligence needed to protect user experience, all without interfering with user traffic.</p><p>The system is lightweight, scalable, and operates using metadata only, with built-in confidence metrics to help teams make better-informed decisions.</p><p>Quality of experience is no longer just a performance metric, it’s a strategic asset. In today’s competitive telecom landscape, the providers who succeed will be those who can see service quality not just as it is, but as it <em>will be</em>.</p><p>With DeepQoE, we’re helping operators make that shift from fragmented monitoring to unified, predictive visibility. From reacting to churn to preventing it. And from managing complexity to turning it into a competitive edge.</p><p>📩 If you’re a network operator ready to take control of service quality at scale, we’d love to show you how DeepQoE can make that possible.</p><p>References:</p><ul><li><a href="https://www.ofcom.org.uk/siteassets/resources/documents/phones-telecoms-and-internet/comparing-service-quality/2025/comparing-customer-service-report-2025.pdf?v=397528">https://www.ofcom.org.uk/siteassets/resources/documents/phones-telecoms-and-internet/comparing-service-quality/2025/comparing-customer-service-report-2025.pdf?v=397528</a></li><li><a href="https://www.gsma.com/newsroom/wp-content/uploads//NG.147-v1.0.pdf">https://www.gsma.com/newsroom/wp-content/uploads//NG.147-v1.0.pdf</a></li><li><a href="https://arxiv.org/pdf/2006.16896">“White Paper on Crowdsourced Network and QoE Measurements — Definitions, Use Cases and Challenges (2020). Tobias Hoßfeld and Stefan Wunderer, eds., Würzburg, Germany, March 2020. doi: 10.25972/OPUS-20232.”</a></li><li><a href="https://arxiv.org/html/2406.08564v1?utm_source=chatgpt.com#abstract">https://arxiv.org/html/2406.08564v1?utm_source=chatgpt.com#abstract</a></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9d5975546a7c" width="1" height="1" alt=""><hr><p><a href="https://medium.com/net-ai-insights/unlocking-accurate-qoe-predictions-in-modern-networks-9d5975546a7c">Unlocking Accurate QoE Predictions in Modern Networks</a> was originally published in <a href="https://medium.com/net-ai-insights">Net AI Insights</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Developing Safeguarded AI with ARIA]]></title>
            <link>https://medium.com/net-ai-insights/developing-safeguarded-ai-with-aria-a23859a56988?source=rss-a2d8f7b90894------2</link>
            <guid isPermaLink="false">https://medium.com/p/a23859a56988</guid>
            <dc:creator><![CDATA[Net AI]]></dc:creator>
            <pubDate>Wed, 02 Apr 2025 09:26:24 GMT</pubDate>
            <atom:updated>2025-04-02T09:26:24.033Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1000/1*nS1mXXB9WWCheb_MwlfNFQ.png" /></figure><p>Net AI has been selected as one of a small number of Creators in the Safeguarded AI programme of the UK Advanced Research + Invention Agency (ARIA). We are honoured to be involved in such a prestigious programme that develops the safety standards needed for transformational AI.</p><p>ARIA is dedicated to funding cutting-edge research that will shape the future — addressing critical challenges across the economy, healthcare, and climate change. In the telecom sector, network operators and equipment vendors are racing to reduce energy consumption and the carbon footprint of the continuously expanding network infrastructure. Net AI is proud to have been awarded a £500k research contract by ARIA to play a key role in this transformation.</p><p>As an ARIA Creator, Net AI will pursue breakthroughs in the development of a safeguarded AI that revolutionises how energy is saved in production Radio Access Networks (RANs). We will build the constituents required to design, train, validate, and test AI-based solutions for effective and automated radio element switch-on/off in modern RANs, aiming to remove data sharing barriers and set the stage for AI operation in the safety-critical telecom domain with quantitative safety guarantees.</p><p>Europe’s bold €200 billion pledge to artificial intelligence earlier this month (read more: <a href="https://ec.europa.eu/commission/presscorner/detail/en/ip_25_467">EU Commission Press Release</a>) aligns perfectly with the exciting trajectory of Net AI, who celebrates its 4th year in business. Net AI’s collaboration with the ARIA programme kicks off against this background of pivotal AI investment. CTO Marco Fiore explains:</p><p><em>“The partnership between ARIA and Net AI will develop a Safeguarded AI workflow to drive the economic and environmental sustainability of next-generation mobile networks, tackling some of the most pressing challenges in the global telecommunications ecosystem. The solutions developed in this project will ultimately enable the industry to facilitate safe access to sensitive data to drive innovation, and to effectively manage energy-demanding RANs.”</em></p><h3>We’re fundraising!</h3><p>Our long-term vision is to revolutionise how mobile network resources are optimised and monetised through AI-driven real-time deep insights and predictive actions. We aim to become the ‘go-to provider’ of software that drives energy efficiency and profitability in the next 5 years. To accelerate go-to-market, we are currently in the middle of fundraising. If you are an investor (or know any relevant ones) who wants to join us on our journey, please get in touch.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a23859a56988" width="1" height="1" alt=""><hr><p><a href="https://medium.com/net-ai-insights/developing-safeguarded-ai-with-aria-a23859a56988">Developing Safeguarded AI with ARIA</a> was originally published in <a href="https://medium.com/net-ai-insights">Net AI Insights</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Post-MWC25 Analysis]]></title>
            <link>https://medium.com/net-ai-insights/latest-from-mwc25-56c2ad968b79?source=rss-a2d8f7b90894------2</link>
            <guid isPermaLink="false">https://medium.com/p/56c2ad968b79</guid>
            <category><![CDATA[mwc-2025]]></category>
            <category><![CDATA[energy-efficiency]]></category>
            <category><![CDATA[ai]]></category>
            <dc:creator><![CDATA[Net AI]]></dc:creator>
            <pubDate>Fri, 14 Mar 2025 12:18:47 GMT</pubDate>
            <atom:updated>2025-03-14T12:26:48.427Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*nvFyk70Y87VGu2F3_y2n9g.jpeg" /></figure><p>As we leave Barcelona behind, we can’t help but reflect on the most relevant conversations that shaped the Mobile World Congress this year. From AI to advancements in non-terrestrial networks (NTNs) and the growing focus on energy efficiency, MWC 2025 remains a prime venue where to take the pulse on the direction in which the telecom industry is heading.</p><p>Here are the key highlights from this year’s event from our perspective.</p><h3>AI Has Taken Roots — But There’s More to Be Explored</h3><p>Much like last year, AI was talked about almost everywhere. Telcos are actively exploring AI adoption, particularly as a way to drive new revenue streams through AI-powered services. However, rather than contemplating AI-powered network management and fully autonomous networks, most conversations revolved around how GenAI and Large Language Models (LLMs) can assist engineers rather than replace them. The hesitation around full automation remains anchored in concerns about model reliability, data security, and the challenges of integrating AI into legacy infrastructure.</p><p>At Net AI, we believe that full automation of a broad set of run-of-the-mill tasks remains a great and vastly untapped opportunity that can improve the way mobile networks operate, relieve network engineers from time-consuming tasks, and ensure significant savings for network providers. Yet, achieving full automation requires shifting attention towards “task- and need-driven” AI/ML model design — one that focuses on solving pressing needs within telco operations, in a reliable and scalable way.</p><h3>Digital Twins: Bringing Networks to Life in a Virtual World</h3><p>Another theme in focus this year was Digital Twins, a concept that is gaining traction in telecom. With network complexity increasing, Digital Twins are becoming an appealing tool — not just for troubleshooting but for long-term strategic planning. By creating virtual replicas of physical networks, operators can simulate, test, and optimize network performance before making real-world changes. This is particularly useful for validating predictive maintenance, traffic management, and energy optimisation solutions that would enable telcos to take proactive steps towards AI adoption.</p><p>Additionally, Digital Twins can help address safety concerns by providing operators and vendors with a secure, controlled environment for testing AI models without exposing sensitive network data or enacting decisions that may risk degrading service operation.</p><h3>The Rise of NTNs: A Step Closer to Truly Global Connectivity</h3><p>Non-Terrestrial Networks (NTNs) was a topic of intense discussion at MWC this year. GSMA organised the <a href="https://www.youtube.com/watch?v=xv9JM5oBICA"><strong>The future of Satellite &amp; NTN</strong> summit</a> which explored key themes in the evolving NTN and satellite infrastructure landscape, including customer demand and willingness to pay for satellite services, the modernisation of satellite architecture, and strategic investments in deploying 5G NTN. By seamlessly connecting mobile devices to high altitude platform stations (HAPS) and satellite networks in a manner that is compatible with the 3GPP ecosystem, the potential exists to fill connectivity gaps in remote or underserved areas cost-effectively. NTNs could further provide critical communications support during disaster response and for environmental monitoring.</p><p>We’re proud to contribute to this transformation as part of the EU-funded ETHER project, which focuses on the seamless integration of terrestrial and non-terrestrial networks. With a vision of 100% global coverage, ultra-reliable service continuity, and drastically improved energy efficiency, ETHER is paving the way for a more connected world.</p><p>Speaking of energy efficiency…</p><h3>Running AI on CPUs: A Shift Towards Sustainability</h3><p>As the telecom industry grapples with growing energy-related OPEX and with the necessity to meet net zero targets, energy efficiency was naturally a hot topic at MWC. In their <a href="https://youtu.be/NgWKrYSGrsA?si=B3JOnnCFe6ej6M1s">Telecom Energy Efficiency Benchmark</a>, GSMA revealed that progress has been made as operators are relying more on renewables and are sunsetting legacy equipment that entails high energy use.</p><p>On the AI front, while GPUs excel at parallel processing and are essential for AI model training, we’re seeing model inference being run on CPUs — possibly augmented with dedicated AI capabilities — where possible to reduce the energy footprint of RAN tasks. This shift highlights a growing trend: AI adoption must balance performance with sustainability to ensure that innovation doesn’t come at the cost of excessive energy consumption.</p><p>Overall MWC 2025 reinforced a clear message: the telecom industry is entering a new era of AI-driven efficiencies and global connectivity. We’re excited to contribute to this transformation by which telecom networks are becoming more intelligent, resilient, and accessible.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=56c2ad968b79" width="1" height="1" alt=""><hr><p><a href="https://medium.com/net-ai-insights/latest-from-mwc25-56c2ad968b79">Post-MWC25 Analysis</a> was originally published in <a href="https://medium.com/net-ai-insights">Net AI Insights</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Driving Energy Efficiency in Open RAN: Net AI’s Partnership with Digital Catapult]]></title>
            <link>https://medium.com/net-ai-insights/driving-energy-efficiency-in-open-ran-net-ais-partnership-with-digital-catapult-5018e851b403?source=rss-a2d8f7b90894------2</link>
            <guid isPermaLink="false">https://medium.com/p/5018e851b403</guid>
            <category><![CDATA[sustainability]]></category>
            <category><![CDATA[5g]]></category>
            <category><![CDATA[energy-efficiency]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[radio-access-network]]></category>
            <dc:creator><![CDATA[Net AI]]></dc:creator>
            <pubDate>Mon, 02 Sep 2024 11:06:09 GMT</pubDate>
            <atom:updated>2025-03-26T13:34:54.991Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*yEAjUjcuUz_A3MhBdYRhGA.jpeg" /></figure><h3>Open RAN and its benefits</h3><p>Net AI, alongside two other innovators, has partnered with Digital Catapult to develop and test energy efficiency solutions for Open RAN (O-RAN) networks. Open RAN is a modern approach in telecommunications that promotes interoperability by utilizing open interfaces and standards, allowing network components from different vendors to work together seamlessly. By separating hardware and software, O-RAN enables the use of general-purpose hardware, fostering competition, innovation, and ultimately leading to more flexible and cost-efficient network management. Additionally, it significantly lowers entry barriers for the smaller suppliers, allowing them to compete by developing specific, innovative components that can smoothly integrate with existing networks through standardized open interfaces. This approach is particularly critical for the deployment of 5G networks, as it supports more adaptable and scalable deployments, while reducing reliance on a few large vendors and promoting a diverse, competitive telecom ecosystem.</p><p>O-RAN also plays a pivotal role in enhancing energy efficiency. Telecom networks use 90% of operators’ energy budgets, with the Radio Access Network (RAN) responsible for 80% of that consumption. O-RAN’s open architecture allows for dynamic resource allocation and intelligent power management, driven by AI algorithms that optimize energy consumption based on real-time traffic demands. The disaggregated nature of O-RAN permits the selection of more energy-efficient hardware, and its support for network slicing further optimizes resource usage, reducing unnecessary power consumption. These capabilities position O-RAN as a key player in the creation of more energy-efficient and sustainable telecom networks.</p><h3>EnergAIze: Net AI’s Solution for Sustainable 5G Networks</h3><p>To address the growing energy demands of 5G infrastructure, Net AI’s solution, EnergAIze, leverages AI-driven forecasting to optimize energy consumption within these networks. As 5G deployments expand and require more extensive base stations with higher capacity, EnergAIze predicts network usage and adjusts Radio Access Network (RAN) configurations in real-time. This automated control significantly reduces energy consumption, achieving over 50% energy savings while maintaining high levels of service quality. The solution’s compatibility with O-RAN presents a great opportunity for telecom operators to balance energy efficiency with network performance, leading to considerable cost savings.</p><h3>EnergAIze’s Role in the Future of Telecommunications</h3><p>Digital Catapult has recently launched its first SONIC Labs Technology Access Programme, which focuses on addressing energy efficiency challenges in Open RAN operations. This programme unites five innovators, including Net AI, to develop and implement sustainable Open RAN solutions. Together with the University of York, zTouch Networks Inc., G REIGNS, and JET Connectivity, Net AI will aim to enhance the performance and energy efficiency of Open RAN through the application of AI and machine learning. This initiative aims to significantly reduce the carbon footprint of global telecom networks while meeting increasing demands, with a showcase event planned for early 2025 to demonstrate the impact and potential of these pioneering solutions.</p><p>As part of this collaborative effort, Net AI will be given the opportunity to test and refine the EnergAIze solution within the SONIC Labs environment. This will not only validate the solution’s effectiveness in realistic network conditions but also allow for improvements, ensuring that EnergAIze is fully optimized for deployment across a range of telecom networks. The anticipated outcomes include substantial reductions in energy consumption, enhanced operational efficiency, and a contribution to the sustainability goals of the global telecom industry. The final showcase in 2025 will highlight these achievements, positioning Net AI as a leader on the path towards greener, more efficient telecommunications infrastructure.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5018e851b403" width="1" height="1" alt=""><hr><p><a href="https://medium.com/net-ai-insights/driving-energy-efficiency-in-open-ran-net-ais-partnership-with-digital-catapult-5018e851b403">Driving Energy Efficiency in Open RAN: Net AI’s Partnership with Digital Catapult</a> was originally published in <a href="https://medium.com/net-ai-insights">Net AI Insights</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>