Operator AI Investment to Exceed $86bn Over the Next Four Years as ‘Zero Touch’ Becomes the Focus

Image
ai

A new study from Juniper Research, the foremost experts in telecommunications, has found operators will increase their annual investment in AI to $22 billion by 2029; rising from $13 billion this year.

Zero Touch

The report predicts this substantial increase of 62% will be driven by operators’ targets of achieving zero-touch operations within their cellular networks. Zero-touch operations significantly minimise or eliminate human intervention in network operations.

An extract from the new report, Global AI in Cellular Networks Market 2025-2029, is now available as a free download.

Agentic AI in RAN: The Most Immediate ROI Benefit

The research identified agentic AI as an emerging technology; enabling operators to achieve zero-touch operations. Agentic AI systems autonomously make decisions based on defined parameters and execute network functions autonomously; backed by complex problem solving. Juniper Research believes that implementing agentic AI in the radio access network (RAN) will be the area that provides the greatest ROI for operators.

Optimising RAN through AI will substantially reduce latency in network decision making; enabling real-time optimisation. We expect more tier 1 operators to begin implementing agentic AI in RAN, but urge these operators to focus on using lightweight models, with a smaller number of parameters. This ensures that the AI models used can operate at the latency necessary to deliver improvements to customer experience. In doing so, these systems will be able to make autonomous decisions at a much faster pace; enabling operators to move towards zero-touch operations and improving user experience.

Research author Alex Webb remarked: “Despite the significant amount operators will invest, the cost savings arising from the reduced energy consumption enabled by implementing agentic AI in RANs is expected to be a substantial contributor to operators achieving ROI.

Share article