en.Wedoany.com Reported - Mavenir and Red Hat have jointly launched the Integrated AI Platform, which combines Mavenir's AI software platform with Red Hat's enterprise-grade Kubernetes and AI capabilities to create a carrier-grade, on-premises, cloud-native AI infrastructure.
Once the platform is launched, operators can monetize AI in the same way they monetize data, by creating token-based consumption plans, fully controlling pricing and service level agreements (SLAs), and adding charges to customer phone bills.
Mavenir and Red Hat stated that the platform supports three operational models. First, as an enterprise AI platform, operators offer metered access to AI models, computing resources, and AI tools as value-added services to customers beyond core connectivity services. Second, as an AI grid infrastructure, operators can charge customers for hosting AI applications and third-party workloads through the platform. Third, operators can use the platform to provide their own AI products and services to users, including AI assistant services billed on a per-token consumption basis.
The platform is designed for on-premises AI models and small language models, and also offers policy-controlled access options to external frontier models when enterprises require greater computing power.
Bejoy Pankajakshan, Chief Technology and Strategy Officer at Mavenir, said that operators are watching AI revenue flow to hyperscale cloud providers and third-party platforms, while they provide the connectivity that makes it all possible. The Integrated AI Platform developed by Mavenir in collaboration with Red Hat provides operators with the infrastructure to become AI service providers in their own right.
Chris Wright, Chief Technology Officer and Senior Vice President of Global Engineering at Red Hat, added that the collaboration with Mavenir provides an integrated solution based on Red Hat's AI-driven Kubernetes-native foundation, featuring MLOps, vLLM inference, and AgentOps capabilities. This hybrid architecture allows operators to run sovereign local models for most traffic and selectively connect to frontier models for specific tasks, all managed through a unified policy and billing framework.
Strictly controlling access to external frontier models may resonate with potential customers. The recent shift in the AI industry from fixed-rate to consumption-based pricing models recalls the bill shock experienced by end users during the early days of metered mobile data usage. Uber CTO Neppalli Naga stated that the company exhausted its annual AI budget in just a few months. Microsoft is migrating developers from Claude Code to its internal Copilot CLI, a decision that is both financial and strategic. Business Insider reported that early AI adopters Amazon and Walmart have recently made efforts to limit employee use of AI to control costs. OpenAI is considering significant price cuts to compete more aggressively with Anthropic and mitigate bill shock. Based on this evidence, a platform with strict restrictions on frontier AI usage has its place and may channel some value to telecom operators.
This article is compiled by Wedoany. All AI citations must indicate the source as "Wedoany". If there is any infringement or other issues, please notify us promptly, and we will modify or delete it accordingly. Email: news@wedoany.com
