In an industry-relevant keynote at the 18th edition of the Telecom Review Leaders’ Summit, Mounir Ladki, President and CTO, MYCOM OSI, highlighted how generative AI (GenAI) and large language models (LLMs) are revolutionizing telecom operations.
Ladki explained that GenAI can "give a voice to the network" and is pivotal in CSPs' transition from traditional telecom operators to technology-driven enterprises. These tools are becoming integral to software systems and facilitating the development of specialized applications like autonomous networks.
Ladki outlined the transformative journey of AI within the telecom sector, emphasizing its progression through three distinct phases:
- Traditional AI: This phase focuses on processing structured data to achieve specific, measurable outcomes. For example, traditional AI has been instrumental in enabling highly accurate capacity forecasting, allowing telecom operators to optimize network resources and plan for future demands. By leveraging well-defined datasets, traditional AI delivers reliable and predictable results, consequently ensuring operational efficiency.
- Expanded AI: This stage broadens the scope of AI applications by incorporating unstructured inputs, such as text, images, or sensor data, to generate structured outputs. For instance, telecom companies can analyze customer feedback or network signals to predict service quality or identify potential bottlenecks. Expanded AI bridges the gap between raw data and actionable insights and implantation, presenting new, innovative possibilities while maintaining a structured framework relevant to the decision-making process.
- Generative AI: Representing the pinnacle of AI technology, generative AI harnesses vast amounts of unstructured data to produce a wide range of outputs, from text generation to synthetic data creation. While this technology holds immense potential in the creation of innovative solutions—such as virtual assistants, personalized marketing, and dynamic service offerings—it also introduces challenges like hallucinations, where AI generates inaccurate or misleading outputs. Addressing these limitations necessitates validation mechanisms and the continuous refinement of AI models.
The next frontier, according to Ladki, involves "agentic architectures," which enhance LLMs for telecom-specific tasks, driving automation and achieving autonomous networks. He emphasized the importance of such networks in advancing services like slicing and specialized connectivity, as they play a much broader role as decision facilitators, providing semantic understanding.
He opined that the industry's goal is to progress from the current tiered autonomous networks to self-healing, fully autonomous networks capable of transforming business intent into technical implementation.
Ladki highlighted MYCOM OSI's innovative GenAI-backed solutions, AInsights and GenAie, as transformative tools for communication service providers (CSPs). AInsights leverages generative AI to deliver real-time insights into network performance and customer behavior, optimizing resources and enhancing satisfaction. GenAie combines AI and automation to streamline operations, manage configurations, and drive efficiency, allowing CSPs to focus on strategic growth in a competitive market.
He concluded by noting that applications like capacity planning, predictive maintenance, and fault management will unlock significant cost savings and efficiency.
Also Read: Next-Level Network Automation: MYCOM OSI’s Latest Unveiled
Also Read: MYCOM OSI Launches EAA Assurance Cloud Globally on AWS Marketplace
