In a recent conversation with Kuba Smolorz from STL Partners, we explored their latest research on how telcos can navigate the AI era. The insights from their report, "The telecom industry in the AI era: How telcos can follow the money" offer a pragmatic roadmap for operators facing pressure from the boardroom to "do something with AI."
Here's a scenario many of you will recognize: Your CEO hears about AI at an industry event. Hyperscalers are pushing the message hard. Your competitors are talking about their AI initiatives. The pressure cascades down through the organization until everyone receives the same task: "What can we do with AI?"
As Kuba and I discussed, this creates an interesting challenge. Suddenly, the method becomes more important than the outcome. Teams are asked to demonstrate AI implementation rather than solve actual business problems. But as STL Partners' research makes clear, this approach gets things backwards.
STL Partners' analysis reveals a fundamental truth: improving operational efficiency through AI is non-negotiable. According to their AI, analytics and automation (A3) research, AI-driven networks can deliver approximately $8-16 million in annual value for an average telco, roughly 5% of current revenues. Nearly half of this value comes from resource management - optimizing how you use assets, people, and processes.
The key is identifying specific use cases that will improve your processes, save money, and enable growth. Yes, better data quality and improved access to data are important benefits. Yes, top management is demanding AI adoption. But the real question is: what problems are you actually trying to solve?
Through our work with telcos at Polystar, we've identified four core machine learning functionalities that consistently deliver value:
These technologies share a crucial advantage: they suppress noise and help you focus on what matters. This dual benefit allows you to drive automated workflows more confidently while also enriching human decision-making.