• Network faults remain one of the biggest and most costly challenges facing telcos
• Root cause analysis (RCA) processes still require innovation
• Major industry bodies are supporting the AI Telco Troubleshooting Challenge, which aims to ultimately improve RCA
• AT&T believes the challenge can help deliver positive outcomes

A broad group of telco and network technology industry organisations, including ETSI,  GSMA, IEEE GenAINet, ITU and TM Forum, have jointly announced and are supporting The AI Telco Troubleshooting Challenge, which “calls on telecom operators, AI researchers and startups to submit advanced large language models (LLMs) that can perform root cause analysis (RCA) on telecoms network faults,” the partners announced. 

They note that “network faults remain one of the most persistent and costly challenges facing the telecoms industry, leading to significant financial losses each year,” and are hopeful the competition can uncover some innovation in dealing with the persistent issues. 

The new competition invites teams to submit AI models in three categories: 

• Generalisation to New Faults – this will assess the best performing LLMs (large language models) for RCA
• Small Models at the Edge – this will evaluate lightweight edge-deployable models
• Explainability & Reasoning – this will focus on the AI systems that clearly explain their reasoning 

Additional categories will include securing edge-cloud deployments and enabling AI services for application developers.

The partners say the aim of the challenge is to “identify practical solutions that lead to more resilient, efficient and automated networks. The judging criteria for the challenge will evaluate entrants for the accuracy, efficiency, reasoning capability and security considerations of their models.” 

The AI Telco Troubleshooting Challenge is also supported by “headline partners” Huawei, InterDigital, NextGCloud, RelationalAI and xFlowResearch.

Dario Sabella, Chair of ETSI MEC (Multi-access Edge Computing), stated: “This challenge addresses some of the most pressing research questions in our industry, such as model generalisation and the efficiency of edge-based AI. By providing unparalleled access to specialised data and resources, we are accelerating the adoption of telco AI. In particular, efficient small language models, suitable for edge AI deployments, will lower the barriers for the AI adoption, making advanced AI solutions more accessible and impactful across the industry.”

Prof. Merouane Debbah, General Chair of IEEE GenAINet ETI, added: “Large Language Models have become instrumental in the pursuit of autonomous, resilient and adaptive networks,” said. “Through this challenge, we are tackling core research and engineering challenges, such as generalisation to unseen network faults, interpretability and edge-efficient AI, that are vital for making AI-native telecom infrastructures a reality. IEEE GenAINet ETI is proud to support this initiative, which serves as a testbed for future-ready innovations across the global telco ecosystem.”

The initiative builds on recent breakthroughs in applying AI to network operations, leveraging curated datasets such as TeleLogs and benchmarking frameworks developed by GSMA and its partners under the GSMA Open-Telco LLM Benchmarks community, which includes a leaderboard that highlights how various LLMs perform on telco-specific use cases.

“Network faults cost operators millions annually and root cause analysis is a critical pain point for operators,” stated Louis Powell, director of AI Technologies at GSMA. “By harnessing AI models capable of reasoning and diagnosing unseen faults, the industry can dramatically improve reliability and reduce operational costs. Through this challenge, we aim to accelerate the development of LLMs that combine reasoning, efficiency and scalability.”

“We are encouraged by the upside of this challenge after our team at AT&T fine-tuned a 4-billion-parameter small language model that topped all other evaluated models on the GSMA Open-Telco LLM Benchmarks (TeleLogs RCA task), including frontier models such as GPT-5, Claude Sonnet 4.5 and Grok-4,” stated Andy Markus, chief data officer at AT&T. “This challenge has the right mix of an important business problem and a technical opportunity, and we welcome the industry’s collaboration to take it to the next level.”

The AI Telco Troubleshooting Challenge is open for submissions on the 28th November and it closes on 1st February 2026, with the winners announced at a dedicated prize-giving session at MWC26 Barcelona.

- Ray Le Maistre, Editorial Director, TelecomTV