Nvidia's latest State of AI in telecommunications survey reveals that autonomous networks and improved customer service have emerged as the primary revenue drivers for AI investment across the telecom sector, with 90% of operators reporting positive financial impact and 89% planning to increase AI budgets in 2026. The research, conducted across 1,038 respondents from September to November 2025, shows autonomous networks leading ROI rankings at 50%, followed closely by customer service improvements at 41%. This positioning reflects a fundamental shift in how telecoms are deploying AI—moving beyond experimental generative AI applications (which 60% of organisations are now using or assessing, up from 49% in 2024) toward agentic AI systems that can coordinate decisions across domains in real time. The acceleration is substantial: 35% of respondents expect their AI budgets to grow by more than 10% annually, driven by tangible returns from autonomous fault prediction, energy management, and configuration drift correction.
For CX teams, the implications are twofold. First, the validation of customer service as a top-three ROI driver signals that investment in AI-powered support infrastructure is no longer discretionary—it's becoming table stakes for competitive positioning. However, the dominance of autonomous networks in ROI rankings raises a critical question: as telecoms prioritise self-healing, self-optimising infrastructure, will customer service AI receive proportional investment, or will it remain secondary to operational efficiency gains? Second, the shift toward agentic AI rather than generative AI represents a material change in how support systems will function. Unlike generative AI's productivity gains, agentic systems operate autonomously across customer journeys, turning insights into decisions without human delay. This means support teams must prepare for a fundamentally different operational model—one where agents coordinate with autonomous systems rather than simply augmenting human agents with better tools.
The broader context matters here: 88% of organisations remain at autonomy levels 1-3, with generative and agentic AI expected to accelerate movement toward level 5 autonomous networks. For support leaders already running platforms like Agentforce or Freshdesk, this creates both opportunity and urgency. The window to shape how agentic AI integrates with customer-facing operations is narrowing as telecoms lock in architectural decisions around autonomous systems. Teams that treat agentic AI as merely the next iteration of chatbot technology will miss the structural ROI that comes from true autonomous decision-making across customer interactions—the same structural ROI that's driving the 89% budget increase forecast.
Artificial intelligence drives autonomous networks, customer service gains Computer Weekly
Artificial intelligence drives autonomous networks, customer service gains computerweekly.com