The telecommunications (telecoms, telcos) industry continues to invest in artificial intelligence (AI) to provide better service to customers and improve profitability.

Like companies in other industries, telecommunications companies understand their future is an AI-powered one. Embracing AI services helps those companies better serve their customers, increase efficiency and ultimately, improve their bottom lines.

A new IBM Institute for Business Value survey of 300 global telecom leaders found that most communications service providers are assessing and deploying gen AI use cases across multiple business areas.

A 2023 study by Nvidia (link resides outside of IBM.com)¹ found nearly 90% of telecom companies use AI, with 48% in the piloting phase and 41% were actively deploying AI. Most telecom service providers (53%) agree or strongly agree that adopting AI would provide a competitive advantage, according to the Nvidia study.

The telecom industry should invest in the right AI technologies and services. That way, they prepare the organization to take advantage of AI’s full capabilities.

AI is producing several advancements in service delivery.

• Machine learning
• Deep learning
• Generative AI
• Digital twins
• Intelligent automation

Machine learning can help telcos crunch massive amounts of information in datasets, sometimes called big data, to create more actionable insights. Machine learning usually involves human activity to help the system better identify patterns and perform tasks.

It can help those telcos take historical data combined with future forecasts to run preventive and predictive analytics to better make sense of trends and maintain a competitive advantage. For example, it can parse customer data to understand usage patterns and better predict when it needs to increase service delivery.

Deep learning is considered a subset of machine learning, except it requires less human intervention and uses multilayered neural networks to simulate the complex decision-making power of the human brain. Telcos can use deep learning to derive even more insights into their network and customer data.

There are several key use cases for gen AI in telcos, especially those related to the customer experience. Companies can use them to better solve customer issues, create personalized content and brainstorm strategic improvements. natural language processing (NLP), gen AI technologies can help telcos tackle many different tasks that historically required manual work.

Examples include co-pilots for software development (link resides outside of IBM.com)², internal knowledge management for support staff and content generation and personalization for marketing and sales departments.

Digital twins are virtual representations of an object or system, intended to provide companies an opportunity to test changes with a simulation without disrupting service. Many digital twins include real-time to most accurately reflect how the real object or system performs. Telcos can use digital twins to test stresses to their network infrastructure and identify different customer usage patterns.

Intelligent automation combines AI, business process management and Robotic Process Automation (RPA) capabilities to streamline and scale decision-making across organizations.

There are several benefits for telecom companies that provide AI-based services.

• Advanced data and analytics
• Enhanced network operations centers
• Improved network performance
• Greater sales growth
• Stronger customer experience
• Supercharged customer service

AI can improve upon predictive analytics and make it even more powerful. Telecommunications providers need to understand how usage patterns change and to avoid outages and provide the right level of service. AI can power the collection and distribution of valuable insights throughout telco organizations and to their partners.

For example, AI can help telcos identify customers likely to churn because of poor network experience. The IBM Institute for Business Value study of telecoms professionals found that 80% of respondents believe that businesses are already using AI to generate new insights from existing data.

A network operations center (NOC) is the brain of a telecom company. It is the centralized location where the company monitors and manages its networks and systems in real time to prevent disruptions and network failures. It can help improve workflows and resource allocation and capacity planning and reduce potentially fraudulent activities.

AI can streamline network performance in several different ways.

• Operational efficiency: Telecom operators can use AI algorithms and AI models to analyze their overall network infrastructure performance, detect usage patterns and adjust to improve latency. Ultimately, AI helps improve network optimization and reduce operational costs.
• Predictive maintenance: Telcos depend on the uptime of their networks. Using predictive maintenance to identify potential issues with hardware or software systems is an invaluable tool. Telcos can then schedule maintenance at a time when disruptions to service are minimal, therefore minimizing customer churn.
• Automated network management: AI can automate various aspects of network management, such as load balancing, traffic routing and capacity planning. This AI use can optimize network performance based on current and anticipated demand, minimizing downtime and enhancing service reliability.

A McKinsey study (link resides outside of IBM.com)³ found that AI can generate up to a 15% increase in sales conversion and up to 10% in capital expenditure cost savings. Telcos companies can use AI to drive content creation personalization and more targeted messages and media buys, by using the technology to continuously improve future marketing campaigns.

Telcos understand that embedded AI in the customer experience provides several benefits. AI can meet customer needs by providing more personalized services and marketing across the customer journey.

Telco companies can use AI tools to parse large amounts of data to analyze customer behavior and customer engagement. They can provide personalized content that they can use to advertiser to advanced segments.

AI can also improve customer journey maps to identify where prospects are falling off and customers fail to become repeat purchasers. AI can optimize customer touchpoints, so telcos market is more efficient and effective.

Most importantly, AI can help telcos identify potential problems (link resides outside of IBM.com)⁴ in their customers’ network service, solving problems before the customer even notices.

Telcos can monitor how AI technologies are improving the customer experience by tracking key customer satisfaction metrics such as net promoter score (NPS), customer effort score (CES) and customer satisfaction score (CSAT).

Customer service representatives can use large language models to better assist customers during calls. AI-driven call centers can use AI applications such as virtual assistants and AI agents to improve customer engagement to solve more customers problems quicker. That approach increases their efficiency and helps customers get back to their other activities.

They can also offer customers self-service chatbots, or conversational AI assistants, powered by AI, to solve their issues without even needing to speak to a customer support representative. The IBM study found that 53% of respondents were already deploying or optimizing AI for customer service and the remaining 47% were assessing.

While AI provides several valuable benefits for telco companies, there are some inherent challenges as well.

• Managing the initial investment
• Knowing which models to use
• Integrating with legacy systems 
• Skills gaps

Incorporating any new technology requires an investment through technology purchase or license. Organizations should allocate funds to license LLM models and might need to invest in either upskilling or reskilling or hiring new employees. But with the right approach, that investment paying for itself through increased efficiencies across the organization, improved customer experience and more successful customer service.

An EY study (link resides outside of IBM.com)⁵ found that 50% of telecom respondents communicated a struggle to identify the right type of gen AI vendor. There are several high-profile vendors and an increasing number of startups offering customized services to specific industries. That’s why it’s so important to work with the right partner to assess options and plot the right path to a solution that works best for every company.

Many telcos might still use legacy infrastructure that is incompatible with modern AI systems. Integrating AI tools into these older systems might require application modernization and IT infrastructure overhauls, such as introducing the hybrid cloud, which can introduce extra costs.

There might be some initial costs for upgrading those systems. But telcos can anticipate reduced IT costs through the cloud and require fewer upgrades and maintenance and more efficient systems in the future.

Adopting AI transforms the organization in many ways. It requires many, if not all, employees to learn new skills so they can incorporate AI tools into their jobs. But the right training programs can address that inexperience and help employees prepare for the AI-driven future. Respondents to the Institute for Business Value study cited inadequate expertise as one of the top barriers to generative AI adoption.

Telcos that improve their employees’ skills can reduce overall labor costs. One reason is that recruiting new employees tends to be more expensive. Another reason is that employees with enhanced skills can do a better job than those employees who are unable to reap the benefits of AI.

AI is already being incorporated into networks, with a primary focus on reducing capital expenditure, optimizing network performance and providing new revenue opportunities. 

• 5G
• The Internet of Things
• Metaverse and virtual reality

5G’s long rollout promised faster connectivity and the ability to connect more devices through IoT, revolutionizing how customers connect with businesses and each other. 

Telcos that use AI capabilities can improve 5G network management and further optimize these advanced networks through predictive maintenance, enhanced security and quicker rollout. Another major benefit of 5G is its ability to connect multiple devices at once, and AI can help streamline that process and find the quickest path to those connections.

Also, 5G technologies can help power the AI user experience, such as making it easier for customers to get answers from generative AI platforms on their mobile phones. 

The Internet of Things (IoT) creates the possibility of a global network of interconnected devices, driving a wide variety of use cases. For example, a smart refrigerator can use IoT to order food and beverage items when it detects supplies are running low.

 In another example, a smart thermostat can lower the temperature in the winter when occupants are at the office, raising it in time for when they return. These devices get smarter through machine learning and other AI technologies and more powerful through greater rollouts of 5G networks. 

Both technologies have had longer rollouts than previously expected. However, many still believe that the metaverse and virtual and augmented technology will be an important part of the future of communications and entertainment. 

There might be increased strains on telecom networks for people accessing these technologies away on cellular data if either grows in importance. It is crucial for telcos to incorporate advanced AI systems to help handle an increased load on their networks.