The Integrated AI Platform introduced by Mavenir in collaboration with Red Hat is designed to help network operators build AI-driven businesses using a familiar consumption model. The platform allows operators to charge for AI services through token-based plans that can be billed directly through existing subscriber relationships. Operators retain authority over pricing structures, service agreements and model selection, creating a framework that aligns AI services with established telecom business practices.

Built on Red Hat AI and Kubernetes infrastructure powered by Red Hat OpenShift, the platform provides a sovereign-first architecture centered on operator control. The system enables deployment of on-premises models and custom-built small language models while also supporting controlled connectivity to external frontier models when advanced reasoning capabilities are required. Operators determine which model processes each request and manage how those interactions are charged and governed under a unified policy structure.

The platform combines Mavenir’s AI software capabilities with Red Hat’s enterprise AI and container technologies on validated hardware supplied through third-party partners. Its architecture prioritizes on-premises and open-source models for most workloads while using intelligent routing to connect selected requests to external models when necessary. Additional capabilities include AI platform-as-a-service functionality, MLOps support, agentic operations, token metering, billing integration, service assurance and zero trust security controls.

Mavenir structured the platform around three deployment approaches. Operators can offer AI infrastructure and services to enterprises through a metered consumption model, use the platform as the foundation for AI grid deployments that host network-based and third-party workloads, or launch AI-powered products directly for their own subscribers. Across each model, operators maintain control over commercial terms, service delivery and operational management while leveraging a common infrastructure environment.

The company says the platform is intended to create new revenue opportunities through AI consumption plans, improve cost predictability by directing most workloads to on-premises models and maintain data sovereignty by keeping subscriber information and model assets within operator infrastructure during standard operations. The framework also supports contractual service commitments through service assurance capabilities and provides deployment guidance based on Mavenir’s own production implementation.

📊 What This Means (Our Analysis)

This announcement focuses on a challenge many network operators face as AI adoption expands: finding a direct role in the AI value chain beyond providing connectivity. By linking AI consumption to billing models operators already understand, the platform creates a pathway for telecom providers to package, manage and commercialize AI services using existing operational structures.

The broader significance lies in the combination of monetization, governance and infrastructure control within a single environment. The platform’s emphasis on sovereign deployments, controlled model access and integrated charging reflects growing demand for AI systems that balance operational flexibility with business accountability, particularly in environments where reliability, security and service commitments remain central requirements.

📌 Our Take: The next phase of AI adoption may depend as much on sustainable business models as on the technology itself.