Nokia said its AI-RAN with NVIDIA is moving into live networks and closer to commercial deployment. The company also said the platform applies AI in radio processing, supports intelligence near where data is generated, and extends governed AI agents into network operations (via www.telecomtv.com).
Radio processing gets the first cut
Nokia said it is applying AI within the 5G signal-processing chain, including channel estimation, beamforming, multi-user MIMO pairing, and scheduling. The company said its AI-native anyRAN software running on NVIDIA Aerial RAN Computer provides a software-defined foundation for integrating AI-powered radio functions that improve network capacity and performance.
Nokia and NVIDIA formed their partnership one year ago to add NVIDIA-powered commercial-grade AI-RAN products to Nokia’s radio access network portfolio. That work now sits inside the live-network narrative Nokia is pushing for the platform.
Operator trials are the proof points Nokia is citing
Elisa ran voice, mobile data, and video analytics simultaneously on the same AI-RAN platform in its 5G network in Helsinki. Nokia said distributed deployment and the evaluation of customer use cases and commercial AI-RAN services are expected in 2027.
BT is preparing to deploy a live-network trial in Birmingham in 2027. Deutsche Telekom and Nokia are developing AI-based channel estimation to improve network performance and efficiency, and initial laboratory measurements showed typical uplink throughput improvements of 10-15%, with larger gains under more complex radio conditions.
Orange plans to continue testing AI-RAN in both laboratory and real network conditions.
The platform is spreading beyond the radio unit
Nokia said its AI-RAN platform is built around AI-led spectral efficiency, integrated sensing and communications, and localized inference. DeepSig is developing spectrum-sensing software that uses existing radio infrastructure to identify RF threats, interference, and rogue transmitters, while ZaiNar is integrating its positioning software with Nokia Cloud AI-RAN to enable precise network-based location capabilities without changes to connected devices or radio hardware.
At NVIDIA GTC Berlin, Nokia is demonstrating integrated sensing and communications that show how the RAN can detect and understand activity in the physical environment. Nokia and NVIDIA are also integrating NVIDIA AI Enterprise software and NVIDIA Nemotron models and libraries into Nokia’s agentic platform for telco autonomy.
Cloud infrastructure vendors are being pulled in
Dell Technologies, Quanta Cloud Technology, and Supermicro are supporting flexible NVIDIA Aerial RAN Computer configurations for distributed and centralized deployments. Red Hat is bringing AI-RAN technologies onto a common cloud-native platform with Red Hat OpenShift.
Nokia will also showcase data center connectivity, IP gateway, and wide-area networking solutions, including 800G coherent pluggables, at GTC. The company said those networking products help cloud and AI builders scale connectivity across a growing set of data center locations.
Droam Perspective
For mobile network operators evaluating AI-RAN, Nokia is no longer talking only about lab work. The disclosure now includes live-network activity, named operator trials, and a 2027 window for distributed deployment and commercial service evaluation.
The bigger question for buyers is how much of this lands as radio optimization versus a wider stack that ties RAN, edge and data-center transport together. Nokia’s own disclosures suggest it wants both.