NVIDIA Says 89% of Telecom Respondents Value Open Models for AI Strategy

NVIDIA says 89% of surveyed telecom respondents consider open-source models and software important to their AI strategy, citing customization and deployment flexibility.

NVIDIA’s October 6, 2026, telecom-focused article makes the case that operators are adopting open models for more than cost savings. It points to the company’s latest State of AI in Telecommunications report, in which 89% of respondents said open-source models and software are important to their organization’s AI strategy.

The article outlines several uses operators see for open models: adapting systems with network and customer data, inspecting and governing model behavior, and deploying models across public clouds, private infrastructure, or edge environments. Operators may also use locally tuned models to offer services tailored to specific languages, industries, regulations, and data requirements. NVIDIA cites the Nemotron family as one set of open models for reasoning and voice applications, with open weights, training data, and recipes.

The examples include SoftBank Corp., which says it is using NVIDIA Nemotron and other open foundations as part of work on its SoftBank Large Telecom Model for network operations, design, and management. NVIDIA also describes Nemotron 3 Large Telco Model, a 30-billion-parameter model fine-tuned by AdaptKey on open-source telecom datasets for industry tasks such as network configuration and customer incident triage. NVIDIA has released a fine-tuning recipe intended to help operators adapt models to their own data and procedures.

AT&T’s chief data and AI officer, Andy Markus, describes the company’s approach as choosing models according to workload needs for performance, cost, and control, rather than relying on a single model. NVIDIA’s article also points to Indosat Ooredoo Hutchison’s Sahabat-AI family as an example of developing open models for Indonesian language and local needs. The post presents open models as one component of wider production systems, alongside data preparation, orchestration, security, and governance tools.

Industry Impact

More control over telecom AI

Telecommunications and network operations

Overview

For telecom executives, the survey points to broad interest in choosing open models to balance performance, cost, and control rather than relying on a single model. Engineering and product teams may benefit from adapting models to network data, operating procedures, languages, and deployment environments. NVIDIA’s examples show activity spanning network operations, configuration, and customer incident triage. In the near term, operators will need to assess model fit and integration requirements; over time, locally tuned systems could support services tailored to regional and industry needs.

Key Impact Points

Open models give operators flexibility to tune systems with network and customer data and deploy across cloud, private infrastructure, or edge environments.

Telecom-specific fine-tuning could support operational tasks such as network configuration and customer incident triage.

The 89% survey finding signals that open models are an important part of many respondents’ AI strategies, not simply a cost-saving option.

Opportunities

Tailored AI for telecom operations

Telecommunications and network operations

Telecom operators can evaluate fine-tuned open models for specific operational workflows, using their own data and procedures to adapt models for tasks such as network configuration or incident triage. The article also points to potential for locally tailored services that account for language, industry, regulatory, and data requirements. This creates practical work for model adaptation and deployment across public cloud, private infrastructure, or edge environments. Any business case should be validated against workload performance, cost, control, and the supporting needs for data preparation, orchestration, security, and governance; the article does not establish realized financial returns.

Fine-tuned models for telecom workflows
Problem: General models may not reflect an operator’s network data, procedures, or local requirements.
Solution: Fine-tune open models on operator data for tasks such as network configuration and customer incident triage.
Target Users: Telecom operators and teams responsible for network operations and AI.
Business Value: More tailored support for operational workflows and deployment choices suited to operator requirements.
Revenue Potential: Operators may use locally tuned models to support tailored services, though the article gives no revenue estimates.

Reference link

https://blogs.nvidia.com/blog/telecom-operators-open-models

Image Source Link

https://blogs.nvidia.com/wp-content/uploads/2026/01/ces26-nvidialive-open-model-ecosystem-1-1680x1200.jpg

Community Discussion (0)

No discussions yet. Be the first to start the conversation!