AI in Telecom: What’s Holding Teams Back—and How to Move Forward 

Why telecom teams hesitate on AI—and how to turn uncertainty into action.

AI is often described as the future of telecom, but for many leaders, it still feels out of reach. Why? Because the concept is so broad that it’s hard to know where to start—and because past promises haven’t always delivered. 

“In most conversations, AI comes across as this giant, amorphous thing,” explains Zain Admani, Co-founder & VP of Engineering at Inorsa. “Leaders are told to ‘add AI to their workflow,’ but it feels like pulling at a tangled ball of yarn. They don’t know which thread to start with.” 

That confusion is compounded by skepticism. “We’re far enough into the AI boom that many people have already been burned,” Zain says. “They tried a solution that didn’t work, and it’s hard to rebuild trust after that.” 

At Inorsa, we believe the way forward is to make AI tangible. Instead of pitching “AI for everything,” we focus on specific workflows where automation can deliver measurable gains—like validating data across documents or reducing errors that stall projects downstream. “The question shifts from ‘what can AI even do for us?’ to ‘here are three things it can do right now,’” Zain adds. 

Drawing generation has already shown dramatic results. On average, customers see a 60–70% reduction in drafting time, with further improvements as teams get used to the tools. In one case, a customer reported an additional 20–30% drop in drafting time after six months of use—proof that the benefits don’t stop after day one. 

So why has telecom been slower than other sectors to adopt AI? The reality is that telecom’s workflows are uniquely complex and messy. Data isn’t always clean, and teams can’t afford long, disruptive transformation projects. They need solutions that work with real-world data, that can be adopted quickly, and that keep projects moving. 

Zain’s advice: “Stop thinking about AI as all-or-nothing. You don’t need everything 100% automated for it to be valuable. The biggest wins are incremental—step by step, workflow by workflow. It’s best thought of as an assistant that helps reduce repetitive work, while humans remain accountable for the outcome.” 

The bottom line: AI doesn’t have to be intimidating. Start small. Pilot one workflow. Once teams see results in their own hands, adoption gets easier—and the path forward becomes clear.