Telecom operators are increasingly finding that artificial intelligence (AI) delivers tangible financial benefits, according to a new report from Nvidia. The company’s fourth annual “State of AI in Telecommunications” survey, released Tuesday, found that 90% of operators report AI is increasing annual revenue and driving down costs. This marks a shift from earlier stages of AI adoption, where potential benefits were often discussed but difficult to quantify.
The survey, which polled more than 1,000 telecom professionals globally, also revealed that 89% of operators plan to increase AI spending in 2026, a significant jump from the 65% who anticipated increased spending a year ago. This investment is being directed toward a new priority: network automation.
While customer service applications were previously seen as a primary leverage case for AI in telecom, the Nvidia report indicates that network automation is now the leading source of return on investment. Roughly half of respondents identified network automation as the top AI use case driving ROI, surpassing customer service and marketing applications. AI models are being applied to predictive maintenance, traffic optimization, fault detection and spectrum allocation, allowing operators to proactively address issues and improve network efficiency.
“There is a seismic shift underway in the telecom industry driven by AI,” said Sebastian Barros, managing director of Circles, a Singapore-based telecommunications provider, in a statement released with the report. “Communication service providers are converging on a new realization. Their role in society extends beyond moving bits across networks toward moving intelligence across local and regulated infrastructure.”
The focus on network automation is prompting vendors to adjust their strategies. Ericsson recently partnered with Mistral AI to embed advanced AI capabilities directly into telecom operations, focusing on automating troubleshooting, modernizing legacy code, and accelerating next-generation network development. This approach prioritizes intelligence within network management systems rather than consumer-facing generative AI interfaces.
AT&T is also leveraging AI for network optimization with its Geo Modeler tool, which simulates environmental and geographic variables before infrastructure deployment. By digitally modeling build scenarios, AT&T aims to reduce capital expenditures and improve coverage planning accuracy, decisions that involve millions of dollars in investment.
Beyond network automation, internal process optimization is also delivering significant returns. Telecom operators manage complex back-office operations, including billing reconciliation, fraud management, and workforce dispatch. AI systems are being used to automate anomaly detection in billing, streamline ticket routing, and optimize technician scheduling, leading to shorter cycle times and reduced manual intervention. AT&T is deploying autonomous AI agents to reduce fraud and customer wait times, utilizing agents capable of analyzing patterns in real time and adapting to new fraud vectors.
Looking ahead, the industry anticipates a transition toward autonomous networks capable of self-configuring, self-healing, and self-optimizing. The Nvidia survey found that 77% of respondents expect to see AI-native networks launch before the deployment of 6G. However, progress toward full autonomy is expected to be incremental, with legacy infrastructure, fragmented IT stacks, and regulatory constraints posing ongoing challenges. Open source models and software are considered important to the AI strategy of 89% of telcos, according to the report.
Cisco Systems recently launched a new AI networking chip and router designed to compete with offerings from Broadcom and Nvidia, signaling increased competition in the AI-driven networking space, as reported on February 10, 2026.
