AI and Autonomous networks empowering the next digital transformation

In our previous blog, we outlined why telecom companies need to have the concept of zero-touch partnering as a key element of their delayering strategy to ensure they can add new services and capabilities as part of their drive to monetize their network. However, as networks become more and more complex, they are becoming incredibly expensive to plan, run and fix and just adding more resources to fill the gap is not an option in a world where telco margins are already razor thin.

This is why over 150 of greatest and largest TM Forum members are all involved in creating a new AI/AN framework. From the members like Google/Microsoft/AWS who are getting very excited around GenAI, along with members such as TELUS, Rakuten, Verizon, and Orange (to name but a few) who are looking to use AI to automate their networks. They are all driven to deliver an end-to-end solution across nine core areas to use AI and automation for a true zero touch operation.

In this blog we discuss how AI and Autonomous networks can be used to empower the next digital transformation.

Automation is not enough to solve the problem

Many of you will have over the years employed various planning, self-optimizing and service assurance tools to help automate your networks all with varying degrees of success as part of your drive to improve employee productivity. However, all aspects of network management need more than to just rules-based solutions, automation also needs agile, responsive and above all intelligent.

However, just saying we need to add AI to automate all aspects of network throws up a magnitude of questions, let alone trust implications, such as:

• When and where do you deploy your AI?
• How do we explain it in the right way?
• How can we make sure that we've got the right framework in place?
• Have we got the right security considerations?
• Do we have the right business model?
• Do you understand what APIs need to be deployed?
• Do we understand who has access to the data?
• Do we even have the right data?
  Practical Implementation is the key

Why are we getting so excited and what are the key sound aspects you need to think about? Well, TM Forum has now been recognized as the de facto standard for what is required to implement autonomous networks. You may have heard about our Maturity levels, these are a real-world, realistic set of requirements on the path to a fully automated network. Many operators now look at these levels to define their roadmap. Take Orange, for example. Laurent Leboucher, CTO of Orange has said that the company is driving to be TM Forum Certified Level 4 by 2025 in change management.

However, as we have discussed in a previous blog, to drive real network monetization an operator needs to determine how they will expose their network either as a NetCo or SerCo but, to achieve either of these, the management of the network is going to require a practical approach to understanding the roll-out of an AI and Automation. You will need to make sure that you cover 3 key aspects:

1. Governance: Both from an AI governance point of view we need to make sure that you know the AI has no bias. It's not going to cause any harm to people. It's to make sure that the algorithms are performing at the rate that we want it to perform. At the same time, if you think about data governance for data governance, what we need to do with that is around data governance because there is no AI without reliable data.
2. Implementation: We need to think about intent where you tell your network what you want it to do in a customer-centric way rather than a technical specific requirement e.g. I need to camera coverage at a football game. How to operationalize its implementation through AIOps and just as importantly understand your level of automation maturity.
3. Application: If we think around autonomous networks, connectivity as a service becomes very, very important as well thinking about network assurance where closed-loop intelligent fault rectification needs to ensure faults are autonomously discovered and fixed within minutes. All this needs a deep understanding of the network which will require operations to have digital twin of the network.

Discover more about the industry’s first AI/AN framework

TM Forum and over 151 of the leading telco members have been working over the years to create an industry first – the newly launched AI and automation framework. Discover more about our solutions to ignite growth.