Harnessing the power of AI in UK telecoms while mitigating its risks

By Dimitra Simeonidou 15 Nov 24 Reading time: 4 mins

Artificial intelligence (AI) is already beginning to reshape the UK telecoms industry in profound ways, and presents tremendous potential to optimise the performance, security, and management of telecom networks.

However, as a core part of the UK’s critical infrastructure – with telecom networks the backbone of communication, data exchange, and the digital economy – we need to also consider the potential risks that AI can pose to our sector.

While AI can be a force for good in managing the complexity of increasingly expansive network infrastructures and services, it can also open new opportunities for malicious actors to exploit vulnerabilities.

Why AI in telecoms is now a growing necessity

AI is not a new concept for the telecoms industry – we’ve been using it to some extent for many years for specific optimisation and operational tasks. However, we are now starting to deploy AI at scale, leveraging data platforms, softwarisation and the cloud in modern networks. The ongoing rollout of 5G, the increasing demand for data, and the proliferation of connected devices through the Internet of Things (IoT) have made the telecoms industry more complex than ever and AI toolkits are emerging to address such complexity.

AI offers unprecedented opportunities to enhance network performance, automate network management and service delivery, and improve overall customer experience.

For example, AI is being used to analyse vast amounts of real-time data and assess abnormal events, isolating and diagnosing the root cause of issues. AI can even dynamically predict, detect, and resolve potential issues or outages before they impact customers. Meanwhile AI-powered chatbots and virtual assistants are streamlining customer support, offering quicker response times and reducing the burden on human agents.

The risks of AI

It’s important to note that the sheer scale of telecoms networks, combined with their distributed nature, makes them vulnerable to cyberattacks. While AI offers a lot of great opportunities for UK telecoms, it also poses significant risks, particularly as it becomes a tool for malicious actors. 

For example, Nokia recently issued a warning about the growing threat of generative AI and automation being used to launch increasingly sophisticated cyberattacks on mobile operators’ infrastructures. 

Nokia’s research found that the number and frequency of Distributed Denial of Service (DDoS) attacks have increased from a maximum of two per day to over 100 per day in many networks. The growth in DDoS attacks is driven by insecure IoT devices, such as smart refrigerators and smartwatches, along with weak security protections, but it is a cautionary tale for the sector. It also found the most common malware in telecommunication networks is a bot that scans for vulnerable devices with weak encryption, passwords, or design flaws.

This highlights how the rise of AI in the hands of cybercriminals presents a new element of risk for the global telecom networks.

AI-driven tools can automate the process of identifying and exploiting network vulnerabilities, which allows cyberattacks to be executed with greater speed and precision than ever before. This raises the stakes for telecom operators, as traditional security measures struggle to keep up with the evolving tactics of AI-powered attackers.

Successfully integrating AI in telecoms

While AI is poised to revolutionise telecoms, its successful integration depends on careful planning and execution. This means designing AI systems with strong safeguards that protect against cyber threats, prevent data misuse, and mitigate operational failures. 

The key to this is embedding safety, security and privacy considerations directly into AI’s development cycle, rather than treating them as an afterthought.

For example, one of the most transformative applications of AI in telecoms is in network management. Modern telecom networks are incredibly complex, handling billions of data points every second. AI offers telecom operators the ability to manage these networks more efficiently by automating routine tasks, identifying patterns in data that humans might miss, and predicting potential failures before they occur. AI-powered network management can optimise bandwidth allocation, improve call quality, and, by embedding security at the outset, can even enhance cybersecurity by detecting and responding to threats in real time.

It’s also worth highlighting that given the scale and sophistication of AI-driven threats, no single telecom operator can tackle the challenge alone. Collaboration across the UK telecoms industry is essential to share insights, threat intelligence, and best practices. By working together, telecom providers can strengthen their collective defences against AI-powered cyberattacks.

This is a key area of focus for UKTIN’s AI Expert Working Group, which consists of 15 members of industry and academia to identify opportunities in AI telco where the UK can take a lead. Its ArtificiaI Intelligence Future Capability Paper on future AI telecoms capabilities in the UK highlights how the reliability, safety, and security of AI systems is critical to the success of their adoption. 

AI offers tremendous opportunities to enhance the performance, security, and efficiency of UK telecom networks. However, it also presents new challenges that cannot be ignored. By embracing AI-driven innovations while taking proactive steps to mitigate the risks, the UK telecoms industry can harness the power of AI to build a more secure and resilient network infrastructure, safeguarding the critical systems that support the nation’s digital future.

Dimitra Simeonidou / Guest writer

Dimitra Simeonidou is UKTIN Lead for UK Research Capability and director of the Smart Internet Lab, University of Bristol.