logo_header
  • Topics
  • Research & Analysis
  • Features & Opinion
  • Webinars & Podcasts
  • Videos
  • Dtw

TM Forum's Innovation Awards 2026 demonstrate progress towards autonomy

TM Forum's Catalyst Awards at Innovate Asia highlight how the industry is collaborating to scale the use of trusted agentic AI, and notably in networks.

19 Nov 2026
TM Forum's Innovation Awards 2026 demonstrate progress towards autonomy

TM Forum's Innovation Awards 2026 demonstrate progress towards autonomy

The 2026 Catalyst Awards at TM Forum’s Innovate Asia highlighted the shift from automation based on individual AI models towards agentic, closed-loop operations. Winners also reflected the growing use of digital twins, intent, knowledge models and governance in scaling agentic AI use.

In total, eighteen Catalysts were on display at Innovate Asia, with more than 220 participants from over 70 companies. Together they “really demonstrated how we’re pushing boundaries through AI-native approaches, autonomous operations, event-drive architectures, and next generation engineering,” according to Aaron Boasman-Patel, Vice President, Knowledge & Innovation, TM Forum.

Details of the winners are below:

Moonshot Catalyst winners

Autonomous Networks challenge: Agentic NOC: Model-driven risk prevention at scale

Agentic NOC tackles a persistent operational problem: network changes can create hidden risks that conventional alarm filtering fails to expose. The Catalyst combines agentic AI, digital twins, domain models and TM Forum standards to detect risks caused by network drift, assess their likely service and business impact, and recommend or execute remedial action. Deep learning and service-impact analysis help operations teams prioritize risks according to affected customers, traffic and SLA exposure rather than alarm volumes. The project is targeting a 20% reduction in major service outages, demonstrating how autonomous operations can shift NOCs from reactive fault restoration towards proactive risk prevention. [inform.tmforum.org]

Explore Agentic NOC | Read the TM Forum Inform article

Showcase excellence: Autonomy accelerated: Connected intelligence for reliable agentic operations - Phase IV

Autonomy accelerated connects business intent, service management, network operations and customer experience to provide a more holistic approach to autonomous operations. The Catalyst combines agentic AI, AIOps, digital twins, knowledge graphs and closed-loop automation, aligned with TM Forum Open Digital Architecture and Open APIs. One key innovation is connecting sales intent with service intent, allowing CSPs to translate demand into technically validated and dynamically priced services. The team is targeting Level 4 autonomy alongside 90% faster time to market and 99% first-time-right service delivery, while exploring new revenues from 5G AI slices, edge AI and guaranteed connectivity. [tmforum.org], [inform.tmforum.org]

Explore Autonomy accelerated | Read the TM Forum Inform article

Open Innovation Catalyst winners

Innovation leadership: Ontology and digital twin-driven intelligent operations and service delivery

Ontology and digital twin-driven intelligent operations and service delivery explores the increasingly important role that shared models and digital representations of networks can play in telecom operations. By bringing ontology-driven understanding together with digital twins, the project is designed to give automated systems richer context about services, resources and their relationships. That approach can help CSPs move beyond isolated automation towards more intelligent operations and service delivery, where decisions are informed by an end-to-end representation of the operating environment. Its Innovation leadership award highlights the growing importance of digital twins as an enabling layer for more autonomous, AI-driven telecom operations.

Explore the project

Business excellence and Showcase excellence: CoreTwin - Autonomous fault management

CoreTwin – Autonomous fault management picked up two awards, for Business excellence and Showcase excellence. The project applies digital-twin techniques to one of operators’ most important operational challenges: detecting, understanding and resolving network faults. By creating a digital representation of network conditions and combining it with automation, CoreTwin is intended to help CSPs understand the consequences of faults and select appropriate responses without relying on lengthy manual processes. The project illustrates the potential commercial value of autonomous operations: reducing operational effort and service disruption while improving network resilience and customer experience. Its two awards recognize both the business case and the execution of the Catalyst.

Explore CoreTwin

Responsible and trusted innovation: PRISM-AI Phase III

PRISM-AI Phase III combines network reliability, sustainability and agentic AI within a governed autonomous control loop. Specialized agents correlate events spanning power, optical, IP, mobile and edge-cloud environments, while a knowledge graph provides operational context. Recommendations are tested in a digital twin before changes reach the network, with explainable AI, agent identities, policy controls and audit trails designed to preserve trust. The team targets 30–40% lower mean time to repair, 60–70% less alarm noise and 15–25% energy savings across addressable loads. The result is a model for autonomous operations in which reliability, sustainability and governance are addressed together. [tmforum.org], [inform.tmforum.org]

Explore PRISM-AI Phase III | Read the TM Forum Inform article

Responsible and trusted innovation: Vision 2030: Smart campus network autopilot

Vision 2030 applies autonomous-network principles to the smart campus, combining Wi-Fi 7, digital twins and AI-driven operations in a unified platform. An intelligent network agent continually monitors conditions, identifies problems and recommends actions, shifting operations from reactive troubleshooting towards proactive optimization. The project is designed to support demanding educational applications including 4K and VR-based teaching, large-scale research collaboration and data-intensive services while improving resilience and operational efficiency. It also explores how network intelligence can contribute to sustainability and new campus services. Aligned with TM Forum’s Autonomous Networks work and Saudi Arabia’s Vision 2030, it demonstrates a practical vertical application for autonomous networking. [tmforum.org]

Explore Vision 2030

TM Forum asset impact: Agentic operations: Accelerating telecom autonomy

Agentic operations: Accelerating telecom autonomy focuses on turning the emerging concept of agentic AI into an operational model that CSPs can apply as they advance towards higher levels of network autonomy. The project demonstrates how AI agents can take on a greater role in analyzing network conditions, coordinating decisions and supporting automated operational processes. Crucially, this is linked to TM Forum assets rather than being developed as a standalone AI architecture. Its TM Forum asset impact award underscores an important challenge for the industry: ensuring that increasingly autonomous AI agents interoperate with established telecom processes, information models and APIs rather than creating another generation of operational silos.

Explore Agentic operations

Impact award winners

Emerging impact: Trusted multi-agent collaboration for autonomous networks - Phase II

Trusted multi-agent collaboration addresses what happens when autonomous networks depend not upon one AI agent, but ecosystems of agents from different vendors and operational domains. The project extends agent-to-agent collaboration with telecom-specific Authorization-T and Observability-T capabilities, providing fine-grained access control, auditability and end-to-end tracing of agent interactions. That allows operators to monitor decisions as well as latency, token consumption and collaboration effectiveness. By embedding governance and visibility into agent collaboration, the Catalyst aims to make multi-vendor agent ecosystems trustworthy enough for production networks. This creates an important foundation for scaling agentic AI towards Level 4 autonomous network operations. [tmforum.org]

Explore Trusted multi-agent collaboration

Deployment with impact: An experience-centric AN L4 solution providing end-to-end user experience assurance

An experience-centric AN L4 solution providing end-to-end user experience assurance reflects the next stage in autonomous networks: judging automation by the experience delivered to customers, rather than network performance metrics alone. The project focuses on Level 4 autonomy and end-to-end assurance, connecting operational intelligence with user experience so that networks can identify and respond to service degradation more autonomously. Winning the new Deployment with impact category is particularly significant because the award recognizes projects progressing towards real operational implementation. It demonstrates how Catalyst work is increasingly moving beyond technology demonstrations towards autonomous capabilities designed to deliver measurable outcomes within communications service provider environments.

Explore the project