• 2 min read
T2 says AI can lift mobile speeds by 10%
T2 has turned on an AI-based mobile network tuning system at more than 200 base stations, with a claimed internet speed gain of at least 10%.
Image: ITzine
T2 has started deploying an AI-based system that adjusts mobile network parameters in real time for individual subscribers. The operator says the approach delivers an internet speed increase of at least 10%.
Instead of serving every device through a one-size-fits-all setup, the system uses machine learning to decide how a smartphone should be handled at that moment. The algorithms factor in signal quality, device location, current network load, and the smartphone’s own capabilities, then choose which frequency bands should be used for data transmission.
T2 says the effect should be most visible during peak hours, when one cell site is splitting resources across many users and handsets with different modem capabilities.
Recommended reading
Oversight Board warns AI models may curb speech
The technology is already active at more than 200 base stations in Altai Krai and the Altai Republic. The pilot covers Aleysk, Zarinsk, Belokurikha, the villages of Altaiskoye and Kyzyl-Ozek.
By the end of 2026, T2 plans to expand the project to another 1,000 base stations, including sites in Barnaul, Biysk, Rubtsovsk, and Kamen-on-Obi.
For mobile operators, the logic is straightforward: mobile internet performance is no longer constrained only by coverage, but also by how efficiently the network allocates radio resources between subscribers. Ericsson and Nokia have spent several years promoting AI-RAN and similar tools for automated radio network tuning. T2's announcement stands out because it describes a rollout on live base stations, even if it is still limited to one region.
That push is also being driven by traffic growth. According to the Ericsson Mobility Report, global mobile data traffic is expected to grow at double-digit rates in the coming years, forcing operators to squeeze more capacity out of existing infrastructure.
If T2 does reach its stated 1,000-station target by the end of the year, the next test will be whether the same 10% gain holds outside the pilot and in denser urban networks.
AI Editor
Ava covers the rapidly evolving world of artificial intelligence, from foundational models and research labs to the real-world economics of intelligence. With a background in computational linguistics, she cuts through the hype to find out what actually works. She firmly believes that benchmarks are just marketing until reproduced in the wild.
via ITzine
