According to Financial Times News, major UK mobile networks including BT EE, Vodafone, Three, and Virgin Media O2 have committed to using artificial intelligence to block scam calls within the next year. The technology will specifically target “number spoofing” where criminals make it appear they’re calling from legitimate UK numbers used by banks or police. This comes as fraud accounts for over 40% of all reported crime in the UK, with £629 million lost to scams in just the first half of 2025. The new telecoms charter with the government also includes advanced call tracing technology to help police track down scammers. Industry data shows 96% of mobile users decide whether to answer calls based on the number displayed, making spoofing particularly effective for criminals.
Sponsored content — provided for informational and promotional purposes.
Why this took so long
Here’s the thing that’s frustrating about this announcement. Ofcom has been pushing mobile operators to fix this problem for three years. The UK’s telephone network is apparently so outdated that criminals can easily fool the system. Basically, we’re still running on technology that makes it simple for overseas call centers to pretend they’re your bank calling from down the street.
An upgrade to 5G would apparently solve much of this by making it clear when calls originate from abroad. So why hasn’t that happened already? It seems like the economic incentive wasn’t there until fraud became such a massive problem. Now with 17% of all fraud cases coming through telecommunications, and those cases accounting for 29% of total losses because they’re higher-value scams, the math finally makes sense for carriers to invest.
The implementation question
Nick Stapleton from BBC’s Scam Interceptors raises a valid point – why will this take up to a year to implement? He’s not wrong when he says a lot of money will be lost during that time. If the technology exists and the commitment is there, what’s the holdup?
My guess? Coordinating between multiple massive telecom companies is like herding cats. Each has their own systems, timelines, and probably legal teams slowing everything down. Plus there’s likely infrastructure upgrades needed that can’t happen overnight. Still, when you consider that UK banks prevented £682 million of fraud in the same period using their own AI systems, you have to wonder if the mobile carriers are moving fast enough.
The bigger picture
What’s interesting here is how this fits into the broader fraud prevention ecosystem. Banks are already using AI to scan payments and block suspicious transactions, and now telecoms are joining the fight on the front end. It’s basically creating multiple layers of defense.
The government’s new fraud reimbursement rules mean most victims will get their money back, but as Stapleton notes, the psychological harm and stress from being scammed can’t be reimbursed. Preventing these calls from ever reaching people is ultimately the better solution.
And let’s be honest – who even answers unknown numbers anymore? I certainly don’t. But the data shows 96% of people still make decisions based on what pops up on their screen, so spoofing legitimate-looking numbers remains incredibly effective for scammers.
What’s next
This feels like one of those “better late than never” situations. The technology exists, the will (finally) exists, and the economic incentive is definitely there. If this works as promised, it could significantly reduce one of the most common vectors for fraud.
But here’s my question – what happens when scammers adapt? They always do. As AI gets better at detecting scams, scammers will get better at evading detection. It’s an arms race, and we’re just seeing the latest volley from the good guys.
The real test will be whether this system can evolve as quickly as the threats do. For now though, any progress against the constant barrage of scam calls is welcome news. Even if it did take three years of pushing to get here.
