Building tomorrow’s telecommunications network today

So it’s an important part of the mission and we think about the design and the architectures of the network. It’s really not even for the next three years. We think about the next 20 and 50 years. Network investments take a long time, and we want to make those investments with economy in mind, but also ensuring the most reliable network offering.

Laurel: You mentioned artificial intelligence and machine learning in a previous answer. In what ways is AT&T using AI and ML, or planning to deploy artificial intelligence?

Raj: Excellent question and also very timely. As a company, we have researchers who have been working on AI for many years. With the advent of much more computing power and much finer data, the opportunity has really opened up over the last, I would say, five years. He plays a very important role at AT&T. Again, we approached AI in an evolutionary way on how we infuse it.

First, we see AI as the engine, and the fuel is data. It starts with how we want to collect data and learn from it. This is where a lot of the machine learning capabilities come in. We’ve invested in a lot of big data management capabilities over the past few years, making sure they’re well exposed to our AI engines. Our data manager, in particular, has worked very hard to establish a democratized ecosystem for data and AI capabilities. There’s a step function here in complexity as the amount of data increases, especially with 5G, and we get a kind of finer grain visibility, and we have a lot more smart controls to then apply the decisions. So we’re taking those steps in this evolutionary way.

Internally, we have many use cases, including how we can use AI for planning, functions, AI for design decisions, but also in real time to help our customers, as well as the network, in various scenarios to deliver better efficiency, better customer experiences, detect security threats, threat analysis, as well as how to use feedback loops to continuously optimize the network. So many use cases throughout the lifecycle.

Laurel: I’m talking about the focus on safety, which is a priority for most executives these days. But not only security, AI and automation also play this very important role for 5G functionality. What other ways does this come into play right now with 5G capabilities?

Raj: Again, this is very timely and a very active area of ​​work. Let me give you some background on how we are structured. Thinking of 5G, we think of it as day zero, day one, day two. Day zero is for planning and forecasting activities. I can see natural ways where AI and machine learning can help you with your predictions. There’s your first day, which is actually building and designing your network. You want to make the most efficiency. Again, feedback loops and reinforcement learning help you do that, as well as using deep learning technology to analyze maps and geospatial data, to figure out where you want to have buried optical fibers and where you want to place a small cell versus a macro cell. So there’s a lot of building engineering where we rely heavily on AI, deep learning, and neural networks.

Then there is a cycle of life, which we call the second day. There are opportunities in there, things like energy savings where we try to optimize the energy footprint of our equipment. Here again, both a business priority, but also a societal priority on the carbon footprint. We see great opportunities for the economy but also to help the planet.