Nokia opens Riyadh R&D centre for AI-native network automation

Nokia has opened a research and development centre in Riyadh, announced on the sidelines of LEAP 2026, to build software that lets communications networks configure and repair themselves with less human intervention. The centre's teams will work on service management and orchestration, self-organising networks, autopilot capabilities, and rApps for automation and energy efficiency. Nokia says the software will target networks that can self-configure, self-heal, optimise their own performance, reduce energy consumption and operate with greater autonomy — and that the centre will also explore AI-native 6G technologies and contribute what it calls Made in Saudi software to Nokia customers worldwide. The distinction between an R&D centre and a sales office is the whole story here, and it is not a matter of degree. A sales office sells products designed elsewhere and adapts commercial terms to a local market. A development centre employs engineers who write code that ships in the global product. The first creates jobs; the second creates capability, and capability is considerably harder to acquire and considerably harder to lose. It matters for a second reason. Software written by people who work in a market tends to reflect that market's problems. Saudi mobile networks operate under conditions that are unusual by European or North American standards: extreme summer heat that changes how equipment behaves and what it costs to cool, enormous seasonal population movements around Hajj and Umrah that no capacity plan can smooth, very high data consumption per user, and a national build-out happening faster than in mature markets. Automation software developed against those conditions solves a set of problems that software developed in Espoo or Dallas would not naturally prioritise. The acronyms are worth unpacking once. Service management and orchestration, or SMO, is the layer that sits above the radio network and coordinates it — deciding what should be configured where, and pushing those decisions down. Self-organising networks, or SON, is the older discipline of letting cells tune themselves and their relationships with neighbours rather than having an engineer set every parameter. An rApp is an application that plugs into that orchestration layer to perform a specific optimisation job. Autopilot is Nokia's framing for running those loops without an operator approving each step. Energy is one of the clearest motivations. Radio networks account for the majority of a mobile operator's electricity consumption, and much of it is spent keeping capacity available for traffic that is not there. Software that can retire and restore capacity safely in response to real demand is one of the few network optimisations with a directly measurable financial return, which is why it appears in almost every automation announcement in the sector. Nokia's stated commercial rationale is the size of the market it is investing into: a Saudi telecoms sector it puts at SAR 84 billion, with high 5G speeds and monthly mobile data consumption of around 53 gigabytes per person — a figure that is high by any international comparison and that puts sustained pressure on radio capacity. The centre lands in a market where the same problem is being attacked from several directions at once. Nokia already has an MoU with Mobily on AI-powered network efficiency and sustainability , while Mobily has separately signed with Ericsson to pursue Level 4 network autonomy using rApps on Ericsson's automation platform . Having a local engineering centre is a competitive argument in that contest: it is easier to promise responsiveness to an operator's specific problem when the engineers are in the same city. The 6G reference should be read as a research direction rather than a product line. Standards work is at an early stage and commercial deployment is years away; what the mention signals is that the centre is intended to do forward research rather than only maintain current products. What Nokia has not published is the size of the commitment — no headcount, no investment figure, no hiring timeline. Those numbers determine whether this is a team of a few dozen or a centre of several hundred, and they are the difference between a presence and an investment. The exportable test is simpler: whether a Nokia customer outside Saudi Arabia ends up running software written in Riyadh. TechScoop is reporting from LEAP 2026 in Riyadh.