As a traditionally tech-forward industry, telcos have leveraged machine learning and predictive artificial intelligence (AI) for over a decade, particularly in network operations. The emergence of generative AI (GenAI) is now accelerating this evolution, presenting both significant opportunities and complex challenges for the industry.

The World Economic Forum has highlighted this trend in its white paper titled “Artificial Intelligence in Telecommunications”, released in February 2025. According to the paper, AI is poised to transform the telecommunications sector, unlocking business opportunities, enhancing operational efficiency and delivering a truly customer-centric experience. tele.net provides a snapshot of the key highlights from the paper…

Four key imperatives

The report projects data consumption to increase from 3.4 million petabytes (PB) in 2022 to 9.7 million PB by 2027. Meanwhile, the telecom industry’s revenues reached $1.6 trillion in 2024. This was propelled by integrated communication service providers (CSPs) that manage network operations and customer-facing services. An increasing number of network companies and service companies are also driving the commoditisation of communication, which has led to stagnating global revenues. Other challenges facing the industry entail a lack of core service differentiation and the need to diversify to drive growth avenues.

At present, CSPs’ operational costs remain stubbornly high, pegged at 65-70 per cent of revenue. It is estimated that network operations alone will account for 50 per cent of the total operating expenses by 2027. Nearly two-thirds of AI professionals across CSPs and hardware or software providers have reported cost savings from AI use cases, with GenAI expected to enhance these efficiencies through data democratisation and automation of repetitive tasks. So, the first imperative use case is the emergence of a new communications model, where AI-enabled network automation management and technology reduce the cost-to-serve.

GenAI capabilities are driving business growth. In business-to-consumer (B2C) setups, AI is reshaping customer acquisition and retention strategies by allowing sales and marketing personalisation powered by predictive models that anticipate individual behaviour. From identifying the purchase propensity of customers and churning risk to providing tailored offerings through granular customer segmentation, CSPs are leveraging multiple B2C strategies. On the business-to-business front, CSPs are increasingly moving up the value chain by capitalising on their infrastructure provision and AI capability development. CSPs are also unlocking growth by developing internally developed AI-as-a-service offerings for small- and medium-sized businesses.

Only 34 per cent of telecom users are satisfied with their service, and 70 per cent are perturbed by a lack of options, despite widespread commoditisation. While traditional AI is already assessing user data and recommending “next best actions” for improving customer interactions, GenAI is transforming these capabilities with advanced conversational tools. These include natural language chatbots and AI-powered retail assistants that provide seamless, personalised interactions by adapting to changing customer needs.

Lastly, a complicated attack surface is produced by AI’s dependence on massive datasets in large-scale computing and storage infrastructures. CSPs are now leveraging AI for secure and reliable operations by automatically identifying and mending vulnerabilities, responding to incidents and preventing fraud. Network planning, adversarial testing, model evaluation, pattern matching and user behaviour analysis are some of the key threat mitigation strategies for which advanced AI is being deployed.

Current industry adoption

Telcos are actively deploying AI across a range of functions to drive efficiency, personalisation and innovation. In sales and marketing, AI enables hyper-personalised customer journeys and content, with GenAI assisting in content creation and lead enrichment. Tools like Telefonica’s “Next Best Action AI Brain” leverage customer data for tailored product recommendations, while AI-powered assistants support sales agents with scripts and solutions.

In products and services, GenAI allows for real-time service configuration and the creation of customised packages for business clients, while predictive models refine segmentation, pricing and bundling. CSPs are also offering white-labelled AI services, such as chatbots and security solutions, to other businesses, and have seen significant improvements in service-level compliance and operational efficiency. A global telecommunications leader implemented “intent-based operations” utilising predictive AI, machine reasoning and complex automation for Digital Nasional Berhad’s Malaysian 5G wholesale network. This system proactively supports 18 diverse service level agreements (SLAs) by identifying and correcting potential breaches, significantly improving SLA compliance from 70 per cent to 100 per cent without compromising other services.

Customer service is being transformed by AI-powered virtual assistants, which deliver human-like interactions and advanced troubleshooting, and by sentiment analysis tools that generate real-time guidance for agents. Live transcript analysis and the automation of administrative tasks optimise service delivery, while predictive insights help agents identify churn risks and upsell opportunities. Notably, AI-driven service automation platforms have led to substantial reductions in service resolution times and improved customer effort scores for European telcos.

In network management, AI is used for digital twin planning, deployment quality assurance, proactive fault prediction, design automation, network decommissioning and energy optimisation. For example, Rakuten Mobile’s AI algorithms reduced detection latency for network issues by over 80 per cent.

Key challenges

For AI to reach its full potential in telecommunications, the industry must overcome several hurdles:

  • Data, infrastructure and architecture: Legacy systems and siloed data remain significant barriers to AI adoption. Modernising technical architecture and building robust data foundations are essential for scaling AI solutions. GenAI’s ability to process unstructured data offers a path forward, but comprehensive data strategies are needed.
  • Workforce, talent and culture: The shift to AI-driven operations requires new skills and a culture that embraces digital transformation. Upskilling the workforce and attracting AI talent are critical priorities. Only 7 per cent of CSPs report being fully satisfied with recent modernisation efforts, highlighting the need for greater organisational agility.
  • Responsible AI: As AI becomes more integral to operations, ensuring ethical, transparent and secure adoption is paramount. Responsible AI principles must be embedded across the organisation, covering everything from data privacy to algorithmic fairness and explainability.

The road ahead

Going forward, AI is set to become the cornerstone of the telecommunications industry’s reinvention. By automating network management, enhancing customer experiences and enabling new business models, AI offers a pathway to sustainable growth and competitiveness. The industry is already exploring transformative AI-driven scenarios such as AI-enhanced data monetisation, sovereign AI infrastructure and fully autonomous networks. CSPs are positioning themselves as providers of AI infrastructure-as-a-service, enabling enterprises to leverage sovereign AI capabilities. GenAI is powering proactive, predictive customer engagement models, and autonomous networks are moving toward “zero touch” operations with self-management and optimisation. The reinvention of the tech stack through agentic AI architectures promises dynamic, context-aware platforms that minimise technical debt and enable continuous improvement, signalling a shift towards a more intelligent, automated and customer-centric telecom landscape.

However, to capitalise on AI’s transformative potential, CSPs must develop a cohesive strategy that fosters collaboration across the value chain. This includes industry partnerships to share best practices and develop common standards; engaging with AI start-ups and established technology firms to co-create innovative solutions; and collaborating with governments and regulators to ensure responsible AI adoption and address societal challenges.