NVIDIA Report: 89% of Telecom Operators See Open Models as Key to AI Strategy

NVIDIA published its State of AI in Telecommunications report on October 6, 2026, and 89% of respondents said open source models and software are important to their company's AI strategy. The report reflects a shift among telecom operators toward building AI on open models for workloads ranging from autonomous networks to customer care.

The company argues that open models give operators five strategic advantages: access to frontier-level intelligence at lower cost, the ability to fine-tune with proprietary network and customer data, greater visibility and control over model artifacts for governance, flexible deployment across public clouds, private infrastructure and edge environments, and the opportunity to host locally adapted AI services for enterprise and government customers.

New Telco-Specific Model and Fine-Tuning Recipe

NVIDIA announced the Nemotron 3 Large Telco Model (LTM), a 30-billion-parameter model fine-tuned by AdaptKey on open source telecom datasets. The model is designed to understand telecom industry terminology and reason through workflows such as network configuration and customer incident triage. NVIDIA also released the full recipe for the end-to-end fine-tuning pipeline, built on NVIDIA NeMo open libraries, so operators can adapt the LTM and other open models to their own networks, customers and procedures.

The Nemotron family of open models provides reasoning performance for agentic workflows and speech capabilities for voice applications, with open weights, training data and recipes. SoftBank Corp. is using open foundations, including Nemotron models, to develop its SoftBank Large Telecom Model for network operations, design and management. Rajeev Koodli, principal fellow of SoftBank Corp. and senior vice president of SB Telecom America, said open models let the company build on global foundation model progress while applying its accumulated network knowledge and operational expertise.

Production Workflows and Local AI Innovation

AT&T's chief data and AI officer, Andy Markus, said the company's approach is not about choosing a single model but matching every workload to the right combination of performance, cost and control. Open models are essential to that strategy, and NVIDIA provides an end-to-end platform powered by NVIDIA AI Enterprise software and NVIDIA Agent Toolkit, spanning data pipelines, open models, agent orchestration, secure runtimes and simulation. Operators need pipelines that anonymize sensitive records and generate privacy-preserving synthetic datasets for fine-tuning, plus a platform that turns open models into governed agentic workflows.

Indosat Ooredoo Hutchison, one of Indonesia's largest telecom operators, is building its Sahabat-AI family of open source models to reflect the country's language and culture. Chirag Sukhadia, chief data and AI officer, said open models let the company adapt intelligence to local language, culture, data and real-world needs, and enable more Indonesians to create with AI on their own terms. Operators across countries can fine-tune open models for local-language and industry-specific needs, and host them on trusted platforms that customers can consume directly or build on with their own data and applications.