UK Mobile Networks Not Ready For AI Traffic
UK mobile networks risk struggling to cope as AI becomes embedded into phones, wearables, vehicles and everyday business applications, a problem recently highlighted by VodafoneThree as it warned that slow 5G upgrades and planning delays could leave Britain’s connectivity infrastructure unable to keep pace with growing demand.
Why Could AI Put More Pressure On Networks?
Much of the infrastructure discussion around AI has focused on data centres, chips and the enormous quantities of electricity needed to run advanced models, but every cloud-based AI service also depends on networks carrying information between those data centres and the people using them.
As AI becomes incorporated into more devices, that traffic could become much more intensive. Phones may continuously send voice, images and other information to AI services, while smart glasses, connected vehicles and business applications could generate frequent exchanges of data rather than simply responding when somebody opens a website or downloads a file.
VodafoneThree has compared the potential effect with the congestion people currently experience when thousands of users gather at a stadium or major event, except that growing AI use could make pressure on network capacity a much more everyday problem.
Why Is The UK Particularly Vulnerable?
Britain has made considerable progress with fixed broadband, with Ofcom reporting that full fibre was available to 82 per cent of UK homes by January 2026 and gigabit-capable connections to 89 per cent. Mobile performance presents a less impressive picture, however, with Which? reporting that the UK ranks last among the G7 for overall mobile-network performance and also behind the EU countries included in its comparison.
The problem isn’t simply whether a phone displays a 4G or 5G symbol, because coverage does not necessarily mean there is enough capacity or reliability to support demanding services consistently.
Standalone 5G is particularly important because, unlike earlier 5G services that still depend partly on 4G infrastructure, it uses a dedicated 5G core designed to provide greater capacity, lower latency and support for more advanced services. Ofcom says standalone 5G coverage outside premises currently ranges from 49 to 85 per cent across the operators that have deployed it.
Planning Is Becoming A Bottleneck
VodafoneThree has committed £11 billion to upgrading its combined network following the merger of Vodafone UK and Three UK, with plans to reach 99 per cent standalone 5G population coverage by 2030. Rather than covering Britain with thousands of extra masts, the company says 96 per cent of the work involves upgrading existing sites and that its overall network footprint will eventually fall by around 30 per cent.
The difficulty is that more than half of those upgrades require some form of planning approval, while approximately 13,500 planned sites require either full planning permission or another approval process before work can proceed.
As VodafoneThree Chief Network Officer Andrea Donà explains on the company’s website: “The good news is mobile operators no longer need to deploy large numbers of new sites to infill coverage gaps, and have broad connectivity footprints across populated areas of the country.”
The operator argues that rules governing changes to mast height and width have not kept pace with modern 5G technology, while differences in how local authorities interpret planning requirements can create further delays.
What Needs To Change?
Planning reform therefore seems to be one of the clearest opportunities, particularly where operators are upgrading existing infrastructure rather than constructing entirely new sites. Making more routine modifications possible under permitted-development rules could allow greater network capacity to be installed without waiting for lengthy individual applications.
The EU has already moved towards tighter deadlines through its Gigabit Infrastructure Act, which generally requires authorities to grant or refuse permits for very-high-capacity networks within four months. As the default position, a permit can be treated as granted if the authority fails to decide within the deadline, although member states can provide alternative remedies.
The UK government seems to have recognised the wider issue. For example, its Mobile Market Review says high-quality connectivity is essential to economic growth and maintains an ambition for all populated areas to have access to standalone 5G by 2030, while examining whether policy and regulation are creating unnecessary barriers to investment.
Improving infrastructure will therefore require more than spending money on equipment. Faster planning decisions, consistent rules between local authorities, continued standalone 5G investment and greater attention to actual network capacity and reliability will all be important if connectivity is to keep pace with new AI services.
What Does This Mean For Your Business?
For businesses, the main lesson is that AI infrastructure extends far beyond the data centre. An organisation can have access to an excellent cloud AI platform, but employees still need fast and reliable connections to reach it, particularly when working remotely or using increasingly data-intensive mobile applications.
Businesses planning greater use of AI should really consider connectivity as part of their adoption strategy, including mobile coverage, broadband resilience and backup connections for sites where reliable access is critical.
The wider UK challenge is similar. Investing heavily in AI models and data centres will deliver limited benefits if the networks connecting businesses and consumers to those services cannot handle the demand, making telecoms infrastructure an increasingly important part of Britain’s ability to compete in an AI-driven economy.
