According to 13F filings, Nvidia just put roughly $1 billion (about 8% of its entire investment portfolio) into Nokia. In Nokia. The company most people remember as the brick phone that survived being dropped down a staircase in 2003.

Jensen Huang didn't spend a billion dollars on nostalgia. There is a specific reason Nvidia planted its flag here. And when you understand what it is, Nokia starts looking like a completely different company.

The financial press covered the filing. They wrote their headlines, moved on, and went back to arguing about data centre margins. Almost nobody stopped to ask the obvious question underneath the obvious story. That question is what we're getting into today.

This MarketBeat piece says the industrial-chip trade is getting fresh momentum because the AI buildout is now spilling beyond GPUs into the boring but essential plumbing: power control, sensing, interfaces, and connectivity. The article’s core idea is that this is not a one-quarter pop. It is part of a broader multi-year upgrade cycle as legacy devices get replaced with more AI-ready systems, which is why names like Texas Instruments are back in focus.

At Berkshire’s annual meeting, Greg Abel flatly rejected the idea of breaking up the company, saying “absolutely not” and stressing that Berkshire still works as an efficient conglomerate. That matters because this was one of the first big public tests of how Abel would frame Berkshire after Buffett, and his answer was basically continuity over reinvention.

The Fool’s point is that Tesla’s robotaxi rollout is still slow, but at least it is now showing measurable progress. The unsupervised fleet grew from nine vehicles in Austin at the start of April to 26 by month-end, with early expansion into Houston and Dallas and preparations underway in more cities, while Europe could also become a sentiment boost if FSD approvals widen. The bigger message is that expectations have come down so much that even incremental progress now looks like real good news.

When you started out, doing your own books made sense. But the business you're running today isn't the one you started. If your accounting hasn't kept pace, it's quietly costing you — outdated financials, no clear view of what's actually profitable, and hours every week pulled away from the work that grows your business. At BELAY, our Financial Experts integrate directly into your business. They manage your books, reconcile accounts, run payroll, and deliver the timely insight you need to make big decisions with confidence. Stop guessing. Start knowing.

Download the Free Guide

Let's be direct about what this story is not. It is not a story about Nokia making a comeback because Nvidia likes the brand. For Nvidia, this investment is explicitly tied to joint product development. That distinction matters enormously to how you read it as an investor.

The obvious read: Nvidia bought Nokia stock, Nokia stock went up, good for Nokia shareholders. Fine. But that's not the real tension here. The real tension is that Nvidia is trying to solve a problem that almost nobody is talking about loudly enough. Where does AI compute go after the data centre? Nokia holds a key part of the answer.

Nokia isn't just selling you better phone signal. It's becoming the physical infrastructure layer through which AI gets delivered to every device on the planet, at the edge, in real time, without routing everything back through a distant data centre. Nokia's installed base of operators and its software architecture is Nvidia's access point into a RAN market expected to exceed $200 billion by 2030.

Here's what that means in practice. RAN stands for Radio Access Network, the part of mobile infrastructure that connects your phone to the broader network. Every base station you see on a rooftop or a tower is RAN. Nokia is one of the world's top three RAN vendors. And Nvidia wants to turn every one of those base stations into a GPU-powered AI compute node.

Nokia's anyRAN architecture gives operators a path from today's conventional networks into AI-RAN without ripping out existing infrastructure. The AI-RAN initiative is positioned as the foundation for a 6G-centric future featuring drones, autonomous vehicles, robots, and extended-reality wearables. If that architecture works at scale, every mobile operator on earth becomes a potential Nvidia GPU customer. Nokia is the distribution network.

The Q1 2026 numbers support the thesis. Revenue of $5.27 billion came in well ahead of the $4.59 billion analyst forecast. Comparable operating profit hit €281 million, up 54% year-over-year. AI and cloud revenue grew 49% and now accounts for 8% of group sales. AI and cloud orders booked reached €1.0 billion in the quarter alone.

And it's not just the numbers. Operators now testing the platform include BT, Elisa, NTT DOCOMO, Vodafone, T-Mobile, Indosat, and SoftBank. That is not a pilot programme. That is a global rollout beginning to take shape.

Nokia is also a company with a division bleeding money.

After setbacks in the US market, Nokia's Mobile Networks business group (the very unit the AI-RAN partnership is meant to reinvigorate) posted an operating loss of €64 million on sales of €5.3 billion across the first nine months of 2025. Between 2023 and the end of 2024, Nokia cut more than 9,000 jobs. It aims to cut another 5,000.

That is not a small detail. The partnership with Nvidia is partly a strategic bet on the future and partly a lifeline for a division under serious competitive pressure.

There's also a deeper technical risk. Ericsson and Samsung, Nokia's two main RAN competitors, have not deviated from their CPU-based, virtual RAN approach. Verizon's CTO has said Intel's Granite Rapids is powerful enough for 6G. Nokia is going all-in on GPU-based RAN with Nvidia. If the industry doesn't follow, Nokia will have bet its mobile division on an architecture that never gets adopted at scale.

The fear is simple: Nokia is spending heavily on a technology that hasn't proven commercial viability at scale, in a division that was already losing money, in a market where its two biggest competitors have chosen a completely different path.

What that looks like in practice: two more quarters of Mobile Networks losses, capex rising toward €1 billion, and operators deciding they'll wait for the technology to mature before committing. Nokia bleeds cash while the market it bet on develops on someone else's timeline. Enterprise tech adoption almost always arrives later than the people selling it expect. That's not a reason to avoid Nokia. It's a reason to know exactly what you're buying.

The easy money on the re-rating trade has likely been made.

NOK has ripped from a Jan 2026 low near $6.06 to around $13.30, a 118% move in roughly three months. The stock is now trading at its highest of $13.30 in the last 10 years. Earnings drove some of it. Analyst upgrades drove more: Bank of America upgraded to Buy with a $12.40 target; CFRA flipped from Hold to Buy and more than doubled its target to $16.

The price is no longer cheap. A P/E above 70 bakes in a lot of optimism. Capex ramping toward €900 million–€1.0 billion in 2026 will weigh on near-term free cash flow. The market has already started pricing in the AI-RAN story.

🧭 A Simple Technical Read

NOK broke out cleanly above multi-year resistance near $10 on heavy volume after earnings. That level matters. The stock had struggled there for years, and clean breaks above long-standing resistance on high volume are one of the more reliable signals that a move has institutional conviction behind it, not just retail momentum.

Watch what happens on the first meaningful pullback. If $10 holds as support (sellers who once dominated at that level now becoming buyers) the breakout has legs and the re-rating is likely sticky. If $10 fails on a retest, the move was event-driven noise and the stock drifts back into its old range. That's your line in the sand. Everything else is commentary.

Three things to monitor, but I'll be honest about which one actually matters most right now.

💰 Mobile Networks Division Profitability: This Is The One

Everything else is secondary to this. The AI-RAN story is exciting. The Mobile Networks operating loss is the uncomfortable reality sitting underneath it. If AI-RAN revenue starts flowing meaningfully through that segment, you will see it in the margins. That division moving toward breakeven in the next two quarterly reports is the single most honest signal that the strategy is working and not just being announced. If margins stay flat or deteriorate while management keeps talking about transformational partnerships, the partnership is still in early innings with no commercial payoff yet. Watch this number every quarter. It is the only one that doesn't lie.

🏁 Competitor Response from Ericsson and Samsung

Right now, Nokia's two main RAN competitors are sticking with CPU-based architectures. This is a binary read: if either Ericsson or Samsung announces a GPU pivot, it validates the AI-RAN market and removes the biggest structural risk to Nokia's thesis. If they double down on CPU (and Verizon's CTO is already on record saying Intel's Granite Rapids is powerful enough for 6G), the adoption battle gets significantly harder for Nokia. Watch this as external confirmation of what operators are actually asking for, not what Nokia is building in anticipation.

📊 AI & Cloud Revenue as a Percentage of Total Sales

AI and cloud orders hit €1.0 billion in Q1, representing 8% of group sales. That number needs to grow toward 15–20% of group revenue before the transformation story becomes the financial reality story. Right now it's a promising line item. At 15–20% it becomes a structural shift. Track it quarterly. It's the cleanest external signal of whether this pivot is compounding or plateauing.

Nokia is not the same company it was five years ago. That part is real. The Infinera acquisition added genuine optical networking capability. The Nvidia partnership gives Nokia a credible path into the AI-RAN era. The Q1 numbers (operating profit up 54%, AI and cloud revenue up 49%) show the transformation is generating real results, not just press releases.

But here's what I keep coming back to. The stock has already moved. Hard.

You're not buying Nokia at $4 with an Nvidia option attached. You're buying Nokia at $12+ with a P/E above 70, a Mobile Networks division that was losing money six months ago, and a technology thesis (GPU-based RAN) that the rest of the industry hasn't committed to yet.

The right mindset here is not FOMO. It is patience. If you believe AI-RAN is a real architectural shift (and there are very good reasons to believe it is), then Nokia is worth owning through the volatility. But entry point matters. A pullback toward the $10 breakout level would be a far more compelling place to build a position than chasing a stock that has already re-rated on sentiment alone.

The Nvidia stamp of approval is real validation. Don't confuse validation with a guaranteed outcome. Jensen Huang makes strategic bets. Some of them take years to pay off.

Talk through your reply and get polished, professional text ready to paste. Wispr Flow strips filler, fixes grammar, and formats everything. 89% sent with zero edits. Works everywhere.

Try Wispr Flow free

Every cycle produces a version of this trade. The asset is real. The technology is real. The partner is credible. And by the time most people are reading about it, the price has already done the work of believing.

Nokia at $12 is not a bad company. It may genuinely be a transformed one. But transformed companies and transformed stock prices are two different clocks, and they rarely sync on your preferred schedule. The investors who do well here won't be the ones who spotted the Nvidia filing first. They'll be the ones still watching when everyone else has moved on.