The telecommunications (telco) sector is rapidly evolving as businesses and consumers seek new applications and experiences that 5G has enabled. This new connectivity is fueling exciting progress in autonomous vehicles, driving innovative new applications for industrial internet of things (IoT), and helping us design and build futuristic smart cities. However, the recent emergence of generative artificial intelligence (AI) has the power to completely transform the industry.

It is pushing telcos to rethink where to place investments supporting transformation – in particular with dwindling net promoter scores (NPS) – a key customer-satisfaction metric. New industry entrants in the form of challenger brands are offering feature-rich self-service experiences that favor higher data caps over traditional voice minutes are also adding to the disruption.

The AI-Powered Telco
Pivoting to new business models has typically been an arduous process, often taking years of re- platforming and complex systems integration projects. However, imagine a world where legacy telcos could rebuild from the ground up using generative AI?

The availability of scalable compute, a massive increase in data, and rapid advancement in the areas of large language models (LLMs), all make the idea of an AI-powered telco – once a far-fetched idea – a real possibility. For many regulated companies such as telcos with privately held data, their extensive proprietary databases present an untapped opportunity. The unique data they've amassed over years of operations has never been opened up to generative AI models before – till now. Generative AI is able to understand, read, connect the dots, find similarities and differences across these very large datasets. For example, agents are now empowered to craft intelligent solutions tailored to a telco's subscriber demographics, and run their operations more efficiently in service of the customers.

Here are my top tips for telco leaders who are looking to start transforming with AI:
Tailor-made or Off-the-Rack AI? Telcos Can Have It Both Ways. By deploying off-the-shelf, ready-to-use generative AI models available through platforms like Amazon Bedrock, telcos can expedite transformation initiatives that previously took years to execute. Customer service can be reinvigorated with multilingual customer service chatbots, network teams can now conduct fault analysis, Root Cause Analysis, and optimization quicker; marketing teams can run personalized campaigns at scale, and developers can leverage secure coding assistants to significantly reduce prototyping timelines.

Early movers like Deutsche Telekom have already seen double-digit improvements and answer accuracy by tapping into powerful AI infrastructure they now have access to. To supercharge productivity, the UK's BT Group has deployed a coding assistant that has generated over 200,000 lines of code for its 1,200 developers. And CelcomDigi in Malaysia will develop Bahasa Melayu language algorithms to create solutions like chatbots for its linguistically and culturally diverse customer base.

Telcos with digital sub-brands can also benefit from generative AI. For example, cloud-native, generative AI-enhanced billing and charging solutions can help telco product managers dynamically generate pricing, discounts, and bundled offers based on fine-grained user segmentation, with the necessary marketing collateral and assets to take these out to their subscribers.
Telcos also have the option of creating their own bespoke foundation models (FMs) tailored to their customers, and the broader industry. South Korea's SK Telecom, for example, created TelClaude to drive AI-powered contact centers, improve existing services like spam detection, and develop personal AI assistants.

AI Accuracy Starts with Data Accuracy and Security
Regulated companies like telcos already possess extensive data from market research, networks, devices, and customer records, which generative AI can rapidly analyze to uncover valuable insights that help companies evolve or even reinvent themselves. However, understanding and protecting this data is crucial. Regulated companies already have policies to prevent data misuse in place and are adept at addressing these concerns, enabling faster AI adoption. This experience gives telcos an advantage over organizations that haven't yet addressed data management issues or don't see a path forward with AI.

"We need to ensure that our data is secure and not used by other actors," the head of IT at a leading telco recently told me.
We know our customers care deeply about the provenance of their data, because in a lot of cases, that data is actually their customers' data. With generative AI, there are justifiable concerns around transparency, bias, and hallucinations (AI-
generated responses containing false or misleading information). These must be carefully governed. Just like all other industries, telcos need to implement rigorous testing and monitoring while having a "human in the loop" to ensure AI systems operate reliably and ethically.

When selecting a generative AI service, my advice to telcos is to prioritize solutions that allow bringing third-party models to your data instead of the reverse. This safeguards data privacy and confidentiality across all businesses.

Intelligent Algorithms Require Human Talent
Generative AI represents a departure from traditional automation, where machines execute predefined tasks with mechanical precision. Instead, it empowers humans by augmenting their capabilities, unleashing creative potential, and enhancing productivity in unprecedented ways.

A 2024 AWS-commissioned AI skills APAC survey which found that 97% of IT and Telecommunications 1 To better understand emerging AI usage trends and where workplaces might be headed, AWS commissioned global research firm, Access Partnership, to conduct an Asia-Pacific AI skills survey. Almost 15,000 employees an employers expect to use AI tools by 2028. As this trend to adopt AI accelerates, telco leaders must provide pathways for employees to pursue AI skills training, cultivating the talent necessary to maximize the potential of generative AI tools. This calls for greater collaboration between governments, industries, and educators to help employers implement AI training programs and guide employees in
matching their AI skillsets to the right roles.

Spanish telco Telefónica recognized that training its internal team was essential to closing the cloud- skills gap and advancing its digital transformation to better support its customers. To achieve this, Telefónica worked with AWS to develop a multiyear upskilling plan. Working with AWS Training and Certification teams, Telefónica Tech equipped its teams with the practical cloud skills necessary to fully leverage AWS capabilities.

AI is a Winning Edge
In 2024, telco leaders face stark a choice: maintain the status quo or boldly reinvent their value-creation formula by embracing generative AI. The stakes are high: it's not only the telcos' success that hinges on their ability to digitally transform, the sobering fact that 70% of business transformation efforts fail means that putting in place a measured strategy is increasingly going to be a table stakes issue, as customers' expectations relentlessly evolve.

Jayanth Nagarajan, Head of Telecommunications Industry, Asia Pacific & Japan, AWS