T-Mobile says it deployed 5G Advanced nationwide in 2025, and is now bringing Ericsson’s AI-native Scheduler with Link Adaptation into live commercial traffic. The operator frames the work as a way to improve spectral efficiency, downlink throughput, and customer experience consistency at scale.
Ericsson’s neural-network scheduler and link adaptation in live traffic
Grant Castle, senior vice president of RAN engineering and emerging tech at T-Mobile, said T-Mobile is continuing to push the boundaries of RAN innovation through its collaboration with Ericsson. In T-Mobile’s account, the AI-native Scheduler with Link Adaptation uses a neural network running on Ericsson’s RAN hardware to predict changing radio conditions in real time, with the goal of enhancing spectral efficiency and boosting downlink data rates.
Castle also said the AI-native approach can deliver a more consistent experience for customers at scale. Ericsson’s position in the same set of announcements is that the scaled live-network results match earlier testing across limited geographies and that the software enables faster, more intelligent decision-making in the RAN.
PGA Championship private 5G: edge operations command center and slicing
Separate from the RAN work, T-Mobile installed a private 5G network for the 2026 PGA Championship at Aronimink Golf Club in Newtown Square, Pennsylvania, for the tournament week. The private 5G setup is intended to deliver mainstream broadcast coverage and connect phones and cameras for player footage and analysis for fans using the event app and social media channels. The operator also said it would use its public 5G network and slicing service as part of the sponsorship deal, and manage edge compute hardware and analytics software as part of the arrangement.
T-Mobile’s press materials describe an on-site operations center it calls a “smart operations command center,” intended to give event teams a “single real-time picture” using an interactive 3D interface. The command center is described as monitoring crowd movement, entry points, concessions, and connected devices, with AI used to help identify issues early so teams can respond quickly; T-Mobile referenced potential examples such as “shorter lines, stocked shelves, smoother entry.”
SuperMobile slices for gates, creator uploads, and data-intensive tasks
T-Mobile said its SuperMobile premium enterprise service offers dedicated network slicing with prioritized speed, built-in security, and satellite-to-device connectivity. For the PGA event, T-Mobile said a slice of its public network will be used for fast ticket scanning at the gates and reliable connectivity for staff. It also said photographers and content creators can upload photos and videos using a “photojournalism network slice.”
Across these slices, T-Mobile described “intelligent connectivity” to optimize performance for data-intensive tasks such as video calls and large file transfers. The event is also scheduled with practice rounds and media events beginning Monday, May 11, and the event proper running from May 14 through Sunday, May 17, with around 150,000 fans visiting the site across the week.
What to track: AI scheduling plus edge and slicing as a single offer
For wholesale roaming and enterprise connectivity planners, the tangible takeaway is less about the marketing arc and more about the operating model: T-Mobile is pairing live RAN intelligence work (AI-native scheduling and link adaptation) with enterprise private 5G, edge compute, and network-slice use cases in a major live event setting.
Grant Castle linked the RAN work to T-Mobile’s 5G Advanced nationwide milestone, while the PGA sponsorship materials spell out how slices and an edge command center are meant to support both broadcast workflows and on-site operations. This kind of managed connectivity approach—spanning network slicing and operational visibility—maps closely to how IoT connectivity solutions are commonly deployed in practice.