Nokia (NYSE: NOK) on Wednesday provided fresh details on its artificial intelligence radio access network (AI-RAN) initiative, naming eight operators that are currently testing the platform and touting a measurable performance improvement. The announcement, however, stopped short of revealing any revenue figures, underscoring the gap between technological promise and commercial reality.
The Finnish telecom equipment maker said A1 Group, Chunghwa Telecom, du, e&, Mobily, stc, TPG Telecom, and Zain Saudi are evaluating the AI-RAN platform through proofs of concept or live trials. Nokia also reported a more than 20% improvement in spectral efficiency — a key metric that measures how much data can be transmitted over a given amount of wireless spectrum.
Shares of Nokia's New York-listed stock rose 3.10% to $10.145 by 11:37 a.m. ET, according to Yahoo Finance data. The timing of the announcement likely contributed to the move, though it's not the sole driver.
Expanding the Trial Funnel
Wednesday's update builds on Nokia's initial AI-RAN platform launch in July, where it first disclosed the greater-than-20% efficiency gain. The company had already announced evaluations with T-Mobile, SoftBank, and Indosat Ooredoo Hutchison. The new roster expands the visible pipeline across Europe, Asia-Pacific, and the Middle East, with Nokia noting that North American engagements are also underway.
While the eight-name list adds credibility to Nokia's AI-RAN narrative, the company did not specify which operators generated the 20% result, the spectrum bands or traffic conditions used, or the cost of the NVIDIA (NASDAQ: NVDA) computing hardware required. It also provided no order values, subscription pricing, contracted site counts, or revenue start dates.
What the 20% Figure Means — and Doesn't
The spectral efficiency improvement is significant because spectrum is both expensive and finite. If software can enable operators to carry 20% more traffic over the same frequencies, they could delay capacity investments or serve more customers with existing assets. That's the economic appeal of AI-RAN, and why a quantified gain is more valuable than a generic partnership announcement.
However, the disclosure is incomplete for valuation purposes. Nokia did not reveal whether the 20% improvement holds up outside controlled trials, nor did it address the power and integration costs of deploying the technology at scale.
Commercial Timeline Remains Distant
Nokia's roadmap, outlined at the July platform launch, targets pilot deployments in late 2026 and commercial availability in 2027 through a software-subscription model. The company also aims for a greater-than-100% spectral efficiency improvement by 2028. Wednesday's result supports the first step but does not validate the longer-term target.
NVIDIA has a real financial stake in the outcome. In October 2025, it agreed to invest $1 billion for a 2.9% stake in Nokia, with the two companies committing to develop 5G and 6G radio software on NVIDIA's architecture. That investment funds development but does not guarantee operator adoption.
Defense MoU Adds Reach, Not Revenue
In a separate release, Nokia announced a memorandum of understanding with UK systems integrator C3IA to collaborate on tactical communications, private 4G/5G networks, command-and-control systems, sensing, autonomous systems, and edge computing for UK defense. Both companies are participants in the UK Ministry of Defence's TacSys procurement framework, which provides a route for public buyers to purchase their services.
The MoU, however, includes no customer orders, minimum purchase commitments, timetables, or contract values. It broadens Nokia's potential addressable market but should not be modeled as backlog.
Investor Takeaway
The bullish case for Nokia is that a multi-region trial roster, combined with a quantified network gain, shortens the path to paid deployments. Operators can test the capacity benefit against their own spectrum costs before committing. If even a portion of the trial partners sign recurring software contracts in 2027, Nokia could add a higher-margin revenue stream without waiting for a full hardware replacement cycle.
The bearish case is equally concrete: trial success may not survive the power, computing, and integration costs of a wide commercial rollout. The next evidence investors need is not another partner name, but a paid order with sites, timing, and economics — followed by proof that subscription revenue improves Mobile Infrastructure margins rather than merely shifting spending between Nokia products.
