This Former Smartphone Maker Has Quietly Become a Top AI Stock

Nokia’s AI-RAN technology targets real-time AI inference latency gaps by turning cell towers into “micro data centers,” with management on track for customer trials later this year (10 customers committed). While Nokia reported only 4% YoY revenue growth in Q1 and margins below 3% for several quarters, its small AI/cloud segment grew 49% and Dell’Oro estimates cumulative AI-RAN spending could reach $35B in five years. Nvidia’s $1B investment in Nokia (part of a partnership for AI-native 5G-Advanced and 6G upgrades) is a bullish signal, and the article notes investors are front-loading positions ahead of AI-RAN commercialization.

Analysis

The market is likely overpaying for the headline optionality and underpricing the implementation risk. AI-RAN is not a software demo; it is a standards, procurement, and carrier-integration cycle, so the monetization lag is more likely measured in quarters than weeks. That matters because the first leg of the trade is sentiment-driven, but the second leg depends on whether trials convert into paid rollouts and whether those deployments carry enough gross margin to change Nokia’s sub-3% profile.

Relative winners are the platform and infrastructure layers with the widest reuse across the AI stack. NVDA has the cleaner asymmetry because it monetizes both the compute layer and the telecom ecosystem pull-through, while Nokia is effectively a higher-beta implementation vehicle with execution risk, customer concentration risk, and a long history of low-quality revenue. If AI inference really shifts to the edge, the second-order beneficiaries are tower/edge infrastructure and backhaul suppliers, while carrier balance sheets could be pressured by higher capex without immediate ARPU uplift.

The contrarian view is that consensus is extrapolating a multi-year TAM into a near-term earnings story. The first 1-3 month catalyst is customer trial news; the 6-18 month question is whether AI-RAN becomes a meaningful budget line or just a PR-friendly overlay on existing 5G refresh cycles. What would falsify the bullish thesis is a delay in trials, weak conversion from the 10 announced customers, or no step-up in margin/guidance by the next two reporting cycles.

Key Decisions for Investors

• Prefer long NVDA vs. short NOK on a 6-12 month horizon: NVDA captures the ecosystem optionality with higher margin durability, while NOK is priced as if trials will quickly become durable revenue; if Nokia converts customers into paid deployments faster than expected, cover the short on any guide-up or margin inflection.
• Do not chase NOK after the 130% rerating; treat it as a show-me story and wait for evidence of paid pilots or order backlog conversion over the next 1-2 quarters before adding risk.
• Watch AMT and SBAC as indirect beneficiaries if AI-RAN drives more edge deployments on towers; a tactical long in tower REITs vs. a short basket of capex-sensitive carriers (VZ/T) makes sense only if carriers confirm higher spending without offsetting monetization.
• If you need convexity, use a small NOK call spread only after first trial-to-contract evidence; before that, the risk/reward is dominated by headline churn and valuation compression once the initial AI enthusiasm fades.