Mavenir and Red Hat Launch AI Platform for Telecom Operators

Purpose-built platform for network operators enables sovereign on-premises AI, operator-built SLMs, policy-governed access to frontier models, and zero trust AI security, all on a single token charging and service assurance platform

Mavenir announced an Integrated AI Platform developed in collaboration with Red Hat that enables network operators to monetize AI the way they already monetize data; through token-based consumption plans, billed on the phone bill, with full control over pricing, service-level agreements (SLAs) and the models that power every interaction. The Integrated AI Platform will operate across three models: operator-branded AI services for their own subscribers; as the AI infrastructure layer powering operator AI grid deployments; and as a managed AI platform through which operators offer AI capabilities to enterprise customers on a token-based consumption basis. Using Red Hat AI as a unified platform that brings together AI capabilities running on Kubernetes infrastructure powered by Red Hat OpenShift, Mavenir’s Integrated AI Platform empowers operators with a flexible, sovereign-first architecture. This allows operators to utilize on-premises models and custom-built small language models (SLMs) that can handle a wide range of use cases, while also benefitting from selective, policy-governed connectivity to frontier models for tasks that demand cutting-edge reasoning capability. Operators control which model handles which request, who pays for it and how it is billed.

Also Read: The AI Architectural Trap: Avoiding One-Way Doors

AI Monetization: The Operator Opportunity

The global AI services market is generating new revenue streams that network operators are uniquely positioned to capture. Token-based AI consumption, the unit model powering every large language model interaction, mirrors the mobile data billing paradigm operators have mastered for decades. Monthly AI token plans, per-department enterprise quotas, and SLA-backed tiers for operator-provided AI services represent a structurally new monetization layer on top of existing connectivity.

To realize this opportunity, operators need production-grade AI infrastructure that controls sovereign on-premises compute, intelligent model routing, token-accurate metering, carrier-grade service assurance, and billing integration. Critically, no single deployment model fits every use case. Routine use cases belong on operator-owned SLMs for cost and sovereignty reasons. Complex reasoning tasks may warrant selective access to frontier models. Built in collaboration with Red Hat, Mavenir’s integrated AI platform supports both modes under a single policy and charging framework, giving operators a hybrid architecture that maximizes economics without sacrificing capability where it matters most.

Also Read: Hyper-Synthetic Data: The Future of Cybersecurity

The Integrated AI Platform

The Integrated AI platform brings together Mavenir’s AI software platform with Red Hat’s enterprise Kubernetes and AI capabilities, delivered on validated hardware from third-party partners. The result is a carrier-grade, cloud-native AI infrastructure platform that operators can deploy on-premises with enhanced data sovereignty. The architecture is designed as hybrid, using default on-premises and open-source models to serve most of the traffic, while Mavenir’s model router provides policy-governed, metered access to external frontier models for use cases that require advanced reasoning or multimodal capability. The platform enables agentic operations and orchestration, intelligent model routing and token optimization, AI platform-as-a-service (PaaS) and MLOps capabilities powered by Red Hat AI; token charging and billing via Mavenir Digital Enablement, closed-loop service assurance, zero trust identity, authentication and AI security, all supported on the carrier-grade container infrastructure of Red Hat OpenShift, the industry’s leading hybrid cloud application platform powered by Kubernetes.

Three Operating Models

The platform supports three distinct operating models, giving operators flexibility in how they deploy and monetize AI:

Enterprise AI Platform: The operator offers metered access to AI models, compute resources and AI tooling to enterprise customers as a value-added service alongside existing connectivity contracts. Enterprise customers deploy their own applications or consume AI APIs on a usage basis. The operator provides infrastructure and platform SLAs; enterprise application SLAs are governed by the specific service agreement between operator and customer.

AI Grid Infrastructure: The operator uses the platform as the compute and AI fabric for AI grid deployments, hosting network-embedded AI applications and third-party workloads on operator-controlled infrastructure. The platform provides infrastructure-level availability guarantees; application performance and service levels are the responsibility of the application owner unless the operator packages the application as its own managed offering.

Operator AI Services: The operator runs the platform to deliver AI-powered products and services to its own subscribers. This includes AI assistant plans billed by token consumption, AI-enhanced network services, and operator-branded AI applications. The operator owns the commercial relationship, sets the SLAs, and retains full control over pricing, branding and service quality.

The platform is designed to deliver tangible outcomes, including:

• New AI revenue streams through bundled token plans from basic consumer tiers to unlimited enterprise quotas.
• Predictable AI economics by shifting the majority of traffic to on-premises models with hard spend caps on any frontier model usage.
• Data sovereignty by design, with all subscriber data and model weights remaining on operator infrastructure under normal operations.
• Contractual SLAs for operator-managed AI services backed by closed-loop service assurance, with platform-level availability guarantees extending to enterprise and AI Grid workloads.
• A rapid path to revenue through pre-validated deployment playbooks, with Mavenir’s own production deployment serving as the reference architecture.
• Enterprise-grade security through zero trust identity controls and AI-specific threat protection that meet the compliance requirements of regulated communications environments.

“Operators are watching AI revenues flow to hyperscalers and third-party platforms while they provide the connectivity that makes it all possible. This changes today. The Integrated AI Platform developed in collaboration between Mavenir and Red Hat gives operators the infrastructure to become AI service providers in their own right. Operators gain sovereign control over models and data, token-accurate monetization that integrates with their existing BSS, and the service assurance to back their own managed AI services with contractual SLAs. Whether operators are launching subscriber AI services, building out their AI grid infrastructure, or offering AI capabilities to enterprise customers on a consumption basis, the platform provides the commercial and operational controls to do it at carrier scale. For the most demanding use cases, the platform also provides policy-governed access to frontier models, so operators are never constrained by what runs on-premises. This is built from the ground up for operator monetization requirements.” Bejoy Pankajakshan, Chief Technology and Strategy Officer, Mavenir.

“In today’s AI-driven world, network operators now more than ever require high availability, data sovereignty and carrier-grade reliability to scale AI across millions of subscribers. Working with Mavenir, we’re delivering an integrated solution supported on the Kubernetes-native foundation powered by Red Hat AI, which brings MLOps, vLLM inference and AgentOps capabilities. This hybrid architecture allows operators to run sovereign on-premises models for the bulk of their traffic and selectively connect to frontier models for tasks that require it, all through a single policy and billing framework. Together, we give operators a path from AI experimentation to AI monetization without rebuilding their operations model.” Chris Wright, Chief Technology Officer and Senior Vice President, Global Engineering, Red Hat.

Write to us [⁠wasim.a@demandmediaagency.com] to learn more about our exclusive editorial packages and programmes.