Telecommunications has always advanced in waves—analog to digital, 3G to 5G, copper to cloud. Today, a new swell is forming at the intersection of TM Forum’s Open Digital Architecture (ODA) and agentic AI. TM Forum’s ODA gives operators a modular, standards-based foundation; agentic AI layers on the autonomous decision support that transform those modules into living, self-optimizing systems. Together, they move the industry from reactive operations to proactive, closed-loop experiences.
Over the past year, Microsoft engineers have road-tested that combination with executives, technicians, customer support representatives, and developers. Regardless of geography or market, operators voiced three universal priorities: break down operational silos, unlock data’s latent value, increase efficiency, and accelerate innovation without eroding trust. At TM Forum DTW Ignite 2025 in Copenhagen, Microsoft is demonstrating how the complementary relationship between ODA and agentic AI converts those ambitions into measurable business outcomes.
Microsoft’s next chapter with the Open Digital Architecture
Microsoft has been a hands-on contributor to TM Forum initiatives for well over two decades, coauthoring Open APIs, chairing working groups, and donating production hardened code that turns standards into deployable solutions. The ODA has become a focal point of that collaboration. By aligning Microsoft Azure cloud-native foundations with ODA’s composable blueprint, Microsoft helps operators assemble best-of-breed solutions without the drag of proprietary silos.
Engineering teams from Microsoft work with communications service providers (CSPs) and industry suppliers to validate specifications, publish reference implementations, and channel field experience back into the standard. The result for operators is faster interoperability, reduced integration cost, and quicker time-to-value for new digital services.
Yet a common obstacle remains: fragmented observability. Every vendor captures telemetry differently, leaving operations teams to deploy ad hoc log aggregators and parsers that inflate costs and slow incident response. Microsoft’s latest ODA contribution addresses this head-on.
- ODA Observability Operator (open source on GitHub)
The operator prescribes a common logging contract, integrates with Azure Monitor, and exposes health data through TM Forum nonfunctional APIs. In early trials, carriers shrank the meantime to detect billing anomalies significantly, freeing teams to focus on proactive optimization rather than forensic log diving. - ODA Landing Zone for Azure
Guidance and a best practice guide on infrastructure-as-code templates that hydrate into an ODA compliant environment—policy, security, and monitoring.
The “Growing B2B with autonomous agents” catalyst project, involving players like Microsoft, Vodafone, and various industry partners, leverages the ODA Accelerator to transform B2B sales for mid-tier enterprise customers by enabling flexible quoting and commerce through generative AI. It enables flexible quoting and commerce, allowing customers to find relevant products using semantic search and create customized solutions that meet their specific business requirements, budgets, and timelines.
These assets illustrate a simple truth: standards only matter when they migrate from documentation into running code. By operationalizing TM Forum guidance, Microsoft accelerates engineering productivity, slashes integration costs, and strengthens the capabilities of telecoms, as well as providing a feedback loop for continual improvement.
Empowering network monetization through network APIs
Through our engagement with CAMARA and GSMA Open Gateway, Microsoft has played a pivotal role in helping operators monetize their networks via a robust partner ecosystem. This ecosystem supports the provisioning, aggregation, and routing of network API requests, enabling seamless integration and enhanced functionality. Our collaboration with industry leaders such as Aduna, Infobip, and Vonage brings aggregated network APIs directly to the Azure Marketplace. This integration grants Microsoft’s global community of developers and enterprises effortless access to essential network functions, including SIM swap detection, phone number verification, real-time device location, and on-demand quality-of-service controls. Standardized through the CAMARA open-source project—co-led by GSMA and the Linux Foundation—these APIs are designed for seamless integration, ensuring that operators can efficiently use network capabilities to drive innovation and growth.
Giving the network a trusted Copilot
Anyone who has joined a major incident conference bridge understands the sense of urgency—and the expense. Multiple teams chase clues, minutes feel like hours, and every second of downtime erodes customer experience and brand equity. Network Operations Agents built with Azure AI Foundry offer another path to successful resolutions. As Cristina Moura Rebelo, Head of AI Community and Ecosystem Engagement at MEO, describes it:
“MEO is transforming into an AI-powered techco, infusing AI into key domain areas and leveraging innovation and technology to create a competitive advantage, business growth, and operational excellency. The first steps made with Azure AI Foundry were key in unlocking the potential of use cases to streamline operations with ChatGOC and the HekaBot, in a scalable, iterative, and agile way, within a very short period of time, delivering outcomes and scaled efficiency to the teams. This is our path to becoming an AI-powered techco.”
—Cristina Moura Rebelo, Head of AI Community and Ecosystem Engagement, MEO
These AI companions ingest real-time telemetry, topology graphs, historical tickets, and vendor manuals; reason over anomalies; then recommend—or even execute—remediation steps under strict guardrails. Every action is logged, policy checked, and auditable so that safety and compliance are part of the operational flow.
At TM Forum DTW Ignite 2025, Microsoft will be presenting on how we are transforming telecom operations with agentic AI, and unveiling the Network Operations Agent Framework, a reference architecture and working pilot environment that operators can explore hands-on. The package includes infrastructure-as-code templates, sample knowledgebase content, and step-by-step guidance for integrating Azure AI Foundry with existing telemetry pipelines. With these assets, communications service providers can progress from proof of concept to production in a matter of weeks—and do so with the assurance that every remediation action remains within corporate risk tolerance.
Unifying data with the Telco Analytics POC Accelerator
Data is the fuel for agentic AI, yet it often sits stranded across disparate clusters, data marts, and line-of-business applications. The Telco Analytics POC Accelerator removes that friction, deploying a domain specific data estate on Microsoft Fabric complete with service assurance, revenue management, and subscriber 360 schemas; lineage policies aligned to data mesh principles; and guidance to connect your backend data sources.
Beyond core ingestion pipelines, the accelerator provides predefined tables for service assurance, revenue management, and subscriber 360, alongside sample queries and dashboards that surface quick wins. Built-in sample data allows developers to prototype AI workloads safely—accelerating experimentation while protecting customer privacy.
When operators gain control of their data estate, they monetize faster, govern better, and feed AI models richer context. Microsoft provides the launch pad.
“Fabric let us build on the familiarity, security, and scalability of Azure. It unites data flows, storage, analytics, and machine learning in a single experience.”
—Jerod Ridge, Director of Data Engineering, Lumen
This unified approach empowers operators to achieve real-time insights and smarter decisions, driving business growth and innovation.
Reimagining business support systems for an agentic world
Business support systems (BSS) are the commercial nerve center of a telco, yet many still feel like 1990s ERP software: dense menus, arcane codes, and labor-intensive workflows. Microsoft’s agentic BSS proof of concept charts a different course.
At its heart is Microsoft Copilot Studio, which leverages TM Forum Open APIs, the Model Context Protocol, and secure tool registration to let AI agents act on behalf of customer care reps. Consider an agent who says, “Upgrade Alessia’s plan to unlimited data and add a family hotspot.” The AI agent validates entitlements, calculates prorated charges, and triggers fulfilment—no swivel chair required. Subscribers upgrade in the time it takes to sip coffee.
Microsoft is equally optimistic about the potential of an Order Fallout Agent. Up to 3% of orders stall in fragmented fulfilment chains. The agent monitors the queue, diagnoses failure patterns, and either self heals or curates a guided fix. In short, the Order Fallout Agent turns a perennial pain point into an autonomous, closed loop process—freeing care agents to focus on higher value conversations and giving customers the seamless experience they expect.
KPN has extended the use of AI companions to their sales operations with Microsoft 365 Copilot. KPN used Microsoft 365 Copilot to enhance their sales operations, streamlining processes, improving customer engagement, and driving business outcomes.
“From the moment a customer contact becomes an opportunity, we link to that information in Microsoft 365 Copilot for Sales, so we can see all relevant data to prepare for a conversation with the customer,”
—Pierrette de Leeuw-Koumans, Lead Generation Team, KPN
Copilot provides real-time data analysis, predictive insights, and automated workflows, enabling the sales team to focus on strategic activities and deliver personalized experiences.
These demonstrations illustrate how BSS complexity can melt away, replaced by conversational experiences powered by open APIs and trustworthy automation. The journey is incremental—operators can start with a single fallout queue or upgrade flow and expand outward.
Momentum stretching from lab to live network
Innovation without adoption is theatre. Microsoft’s ecosystem partners are translating blueprints into operational gains:
- Microsoft and leading BSS suppliers are exploring joint proof of concepts that integrate the Telco Analytics POC Accelerator and Observability Operator into next generation revenue assurance workflows.
- PLDT has implemented the Amdocs Customer Engagement Platform, a robust, telco-grade solution jointly engineered by Amdocs and Microsoft elevate customer experience management. “By combining the AI, generative AI, cloud, and deep telecom expertise of Amdocs and Microsoft, PLDT an end-to-end solution that will drive higher agent productivity, operational efficiency, and significantly improve customer loyalty,” said Anthony Goonetilleke, Group President of Technology and Head of Strategy at Amdocs.
- Nokia’s NetGuard Cybersecurity Dome is providing comprehensive security for 5G networks, leveraging AI and automation to detect, manage, and respond to threats in real-time.
- Accenture, Tech Mahindra, and other global SIs are collaborating with Microsoft on service offerings that accelerate deployment of AI-ready data estates—combining migration expertise, reference architectures, and operator specific best practices.
The breadth of deployments demonstrates that Microsoft’s approach scales across geographies, regulatory regimes, and network generations.
Charting the first step
Building toward autonomous operations seldom begins with a blank slate. The most effective starting point is a business moment that already matters—whether it’s easing congestion at a busy urban cell site or clearing a stubborn order backlog. Instrument that scenario end to end, unify the supporting data, introduce a focused agent, and track the results with discipline. Momentum builds quickly when measurable wins are visible to both engineers and executives.
Microsoft and its partners stand ready to help, whether through co-innovation blueprints, rapid pilots leveraging the ODA Accelerator for Azure, or structured engagements that blend domain expertise with change management.
Telecommunications remains, at its core, a human endeavor: engineers who safeguard critical infrastructure, customer care teams who build loyalty, strategists who spot the next market opportunity. Agentic AI amplifies that expertise—it automates repetitive analysis, highlights hidden insights, and executes well understood actions—while judgment, creativity, and empathy stay firmly in human hands. By pairing people with autonomous assistance, operators can scale excellence without sacrificing the personalized touch that defines great service. Microsoft invites the industry to explore that partnership at TM Forum DTW Ignite 2025 and beyond.
Join the journey
Learn more by visiting the Microsoft Telecommunications Industry hub, where solution briefs, customer stories, and partner offers provide actionable next steps. Together, the industry can turn aspiration into action and chart the next great wave of telecom innovation.
Microsoft for telecommunications
Accelerate telecom transformation in the era of AI
Rick Lievano, Microsoft’s WW CTO for Telecommunications, develops solutions and architectures for service providers. An active industry speaker and advocate for open standards, he is a member of the TM Forum Collaboration Sub-Committee and leads AI initiatives to help operators achieve more.
