- Nokia VP told Fierce that operators are at the start of their automation journey
- He said that use cases like self-optimized networks are prevalent now
- By 2028 to 2030, he expects many operators to be at a high level of autonomy
Nokia’s autonomous network expert told Fierce its operator customers are still at the start of the journey on automation but are aiming for high levels of autonomy by the end of the decade.
Nokia’s VP of Secure and Autonomous Networks Rodrigo Brito said that operators are looking to make their operations more automated “between 2028 and 2030."
He wouldn’t name specific customers Nokia is working with. However, T-Mobile has signed a multi-year RAN deal with the vendor, while AT&T has a core deal in the United States. In Europe, the vendor has deals with Deutsche Telekom, Telefonica and others. In India, the vendor has signed agreements with Bharti Airtel and Jio. So, Nokia has plenty of tier 1 operators worldwide that are likely examining their autonomous options.
Brito hinted that some of his big tier 1 customers have jumped headfirst into the automation game. “The big ones in North America and the big ones in Europe, they are, I would say, quite advanced," he stated.
He added that smaller operators are also spending on automation to enable them to compete. “We also see...tier 2 operators investing significantly on it because they are actually facing limitations in terms of people and available resources,” Brito explained.
Where are we with automation?
Brito said that automation has already started on the RAN level and is also taking place at the core level now. “The low-hanging fruit has always been the RAN...many of the automation use cases started already, a while ago, with self-organized networks,” he said. “The RAN brings the volume but there is also a lot of automation that we are doing on the core and the transport.”
Brito said that Nokia is having success being able to “automatically monitor a large network” with hundreds and thousands of cells and “being able to anticipate a performance drop a few hours ahead of that performance drop happening.” The network, he continued, recognizes that the problem will occur based on “performance KPIs” or “pattern recognition” and is thus able to fix the problem before it happens.
AI agents are also coming into the picture, Brito noted. He said operators are using AI agents to create slices and “to coordinate other AI agents and solve problems.”
Where are we going with automation?
All in all, Brito expects that operators “will be at a high level of autonomy” within a few years.
“The majority of them, I would say, would want to reach...a highly autonomous level, so level 4, with several use cases that can operate without any human in the middle,” he said. “We have the resource level...and then we have the service level...Both will be automated and some services will be completely automated.”
Brito still doubts, however, that many use cases will reach level 5 autonomy.