A message from our friends at Paradigm Press (sponsor) |
Dear Reader, |
In the next 3 minutes... |
James Altucher – legendary investor and venture capitalist... |
And someone who's known for playing his cards "close to the vest"... |
Is going to give you the name and ticker symbol of a company he believes will skyrocket thanks to the coming Starlink IPO... |
100% FREE of charge. |
No credit card or any other info required. |
Click here to watch this short 3-minute video now. |
This is something James almost NEVER does... |
But when Starlink eventually goes public – as soon as March 26th... |
It could create $100 BILLION worth of wealth in a single day. |
And this FREE, pre-IPO play – from one of the world's most successful investors... |
Could be your chance to share in this epic wealth boom. |
Click here to discover it now. |
Regards, |
Matt Insley
Publisher, Paradigm Press |
|
BONUS ARTICLE |
Oracle Just Repriced the AI Buildout — Here's an Under-$20 Power Chip Name Worth Watching |
Sometimes the market tells you a story in one day. |
And sometimes it tells you a story in one quarter. |
This morning, Oracle did both. |
The stock ripped higher after delivering what was, by any reasonable standard, a statement quarter. Fiscal Q3 revenue came in at $17.19 billion, up 22% year over year. Adjusted EPS landed at $1.79. Oracle Cloud Infrastructure revenue jumped 84% to $4.9 billion. Total remaining performance obligations exploded to $553 billion, up 325% from a year ago. And perhaps most important of all, Oracle raised its fiscal 2027 revenue target to $90 billion, above the roughly $86.6 billion Wall Street had been expecting. |
That is not a "nice quarter." |
That is a market signal. |
It says the AI infrastructure race is still real, still capital-intensive, and still starving for one thing more basic than software, more basic than databases, and even more basic than GPUs: |
power. |
And that is where today's Rising Star idea comes in. |
Not another trillion-dollar cloud giant.
Not another obvious GPU winner.
Not another data-center REIT. |
Instead, the stock I would keep on radar here is Navitas Semiconductor (NVTS), which has recently traded around $8 to $9 a share. That keeps it well under your $20 threshold. More important, it sits in a tiny but increasingly strategic corner of the AI stack: high-efficiency power semiconductors for next-generation AI data centers. |
Now, let me be precise here, because precision matters. |
Navitas is not a disclosed major direct Oracle vendor in the way Nvidia is. I do not want to overstate that. What we do know is this: Oracle Cloud Infrastructure has been named by Nvidia as one of the operators designing for 800 VDC data centers, and Navitas has publicly said its GaN and SiC power devices are purpose-built to support Nvidia's next-generation 800 VDC AI factory architecture. That makes Navitas a very credible second-derivative Oracle infrastructure beneficiary if Oracle's AI buildout keeps accelerating. |
That distinction is important. |
Because when Oracle blows the doors off earnings, most investors look at the obvious names first. |
Nvidia. AMD. Broadcom. Maybe Vertiv. |
But when you look one layer deeper, the real question becomes this: |
What happens when hyperscalers stop arguing about AI demand and start solving the engineering nightmare of feeding enough electricity into thousand-GPU clusters without wasting half of it as heat? |
That is the lane Navitas is trying to own. |
Oracle's quarter changed the debate |
For most of the last six months, the Oracle argument was messy. |
Yes, the company had huge AI ambitions.
Yes, it had giant contracts.
Yes, Larry Ellison was talking about a future measured in gigawatts, not just software seats. |
But the skeptics kept circling the same pressure points: capex, debt, execution risk, and whether Oracle had overreached by sprinting into the AI infrastructure land-grab. |
This quarter did not remove every concern, but it did change the tone. |
Oracle said more than 90% of its capacity expansion over the next three years is funded through partners, and management emphasized that many AI customers are either prepaying for equipment or providing the expensive GPUs themselves. That matters, because it suggests the bottleneck is shifting away from demand visibility and toward physical deployment. Oracle's own executives framed the next phase in terms of data-center and power capacity, not just software uptake. |
In plain English: Oracle just told the market that AI is no longer mostly a "who has the better model?" game. |
It is increasingly a who can build, energize, and cool the compute fast enough? game. |
That is a very different investing environment. |
Because once that happens, the winners broaden. |
Suddenly the conversation includes rectifiers, converters, breakers, transformers, optics, cooling loops, storage, and power-management silicon. |
And at the micro level, one of the more interesting enabling pieces is the move from traditional alternating-current designs toward 800 VDC architectures in AI factories. Nvidia has argued that 800 VDC can improve scalability and efficiency, reduce materials usage, and support the far higher rack power requirements that next-generation AI clusters demand. Oracle Cloud Infrastructure was specifically listed among the companies designing for those kinds of data centers. |
That is why Navitas matters. |
Why Navitas is interesting here |
Navitas does not have the cleanest income statement in semiconductors. It is not yet a mature compounder. It is not printing huge revenue numbers. And that is exactly why it still trades like a speculative infrastructure pick rather than a fully discovered AI winner. |
But look at what the company is actually building. |
Navitas has been rolling out 650V GaN, 100V GaN, and high-voltage SiC solutions aimed at AI data-center power architectures. In October, it explicitly tied those products to Nvidia's next-generation 800 VDC AI factory computing platforms. In February, it introduced a 10 kW DC-DC platform for next-gen AI data centers that it said could deliver 98.5% efficiency in 800 VDC and +/-400 V environments. |
That may sound like engineer-speak. |
It is not. |
It is economics. |
At AI scale, tiny efficiency improvements become enormous. |
If you are operating megawatt-scale racks and gigawatt-scale campuses, every lost percentage point in conversion efficiency turns into real money, real thermal stress, and real infrastructure complexity. The whole point of moving toward higher-voltage direct-current designs is to reduce conversion steps, lower resistive losses, and push more usable power into the rack. Nvidia's own materials make exactly that case. |
So when Oracle reports OCI growth of 84% and says it has secured more than 10 gigawatts of power and data-center capacity coming online over the next three years, that is not just bullish for cloud revenue. It is bullish for the entire chain of companies trying to make those gigawatts technically and economically viable. |
Navitas is one of the smaller names in that chain. |
Which is exactly why it deserves attention. |
The numbers: small company, very early, very volatile |
This is where we need to be honest. |
Navitas is not a polished story stock with perfect fundamentals. |
For the fourth quarter of 2025, the company reported revenue of just $7.3 million, down from $18.0 million in the year-earlier quarter. Full-year 2025 revenue came in at $45.9 million, and the company posted a GAAP net loss of about $117.0 million for the year. On the other hand, it ended 2025 with approximately $236.9 million in cash. |
That tells you two things immediately. |
First, this is not a current-earnings story.
Second, this is a company still in the "prove it" stage of commercialization. |
That makes the stock risky. |
But it also explains the valuation. |
NVTS is not priced like a consensus AI essential. It is priced like a company that the market believes might become important if 800 VDC and high-efficiency power conversion become standard in large AI facilities. That is a very different setup from buying a name everyone already agrees is indispensable. |
And that is why the stock tends to move violently on ecosystem news. |
When Nvidia elevated Navitas's role in next-gen AI power delivery last year, the stock surged sharply. Barron's reported a 17% jump on one of those sessions, while Investopedia noted the stock had more than doubled from earlier levels after Nvidia-related enthusiasm. |
That kind of behavior tells you the market already understands the basic possibility: |
If AI factories become constrained by power architecture rather than just chips, small suppliers with the right silicon could rerate hard. |
The hidden insight: Oracle's blowout is really a power-delivery story |
Most people reading Oracle's earnings will stop at cloud growth. |
That is too shallow. |
The more interesting interpretation is that Oracle is dragging the market into the physical layer of AI faster than many investors expected. |
A year ago, it was still acceptable to talk about AI as if it were mostly software plus GPUs. |
Now Oracle is talking in terms of funded capacity, giant RPOs, data-center construction, and power availability. Its own cloud-infrastructure materials discuss the challenge of synchronized GPU power swings stressing transformers, PDUs, and upstream generation systems. Oracle is not just buying compute. It is redesigning the electrical nervous system around compute. |
That matters because it changes what a useful supplier looks like. |
The next great AI winners may not all be the ones making the chips that do the thinking. |
Some will be the companies making sure those chips can be fed, cooled, and connected without breaking the economics of the data center. |
Navitas sits squarely in that thesis. |
Again, to stay disciplined: I am not claiming Oracle has disclosed Navitas as a major purchase order winner. What I am saying is that Oracle is leaning into an architecture transition that directly favors companies working on higher-efficiency power-delivery solutions, and Navitas has publicly positioned itself for exactly that transition through Nvidia's 800 VDC ecosystem. |
That is a subtle distinction. |
It is also the kind of subtle distinction that often creates the best "rising star" opportunities. |
What investors should watch next |
If you are going to track NVTS intelligently, do not just stare at the stock chart. |
Watch four things. |
First, watch whether Oracle's AI capex narrative holds over the next quarter or two. If OCI growth stays elevated and funded capacity keeps coming online, that strengthens the broader case for suppliers tied to next-gen power architecture. |
Second, watch whether Nvidia's 800 VDC roadmap gains wider adoption beyond pilot language and conference-stage enthusiasm. The more that architecture becomes standard, the more valuable Navitas's current positioning becomes. |
Third, watch Navitas for design-win evidence and actual revenue conversion. Product announcements are nice. Sampling updates are nice. But eventually this company needs rising commercial revenue from AI data-center programs, not just promising press releases. |
Fourth, watch the balance sheet. Navitas had cash at year-end, but losses are still large relative to revenue. That means execution matters, and dilution risk can never be ignored in a story this early. |
Bottom line |
Oracle's earnings blowout was not just a win for Oracle. |
It was a wake-up call for the market. |
The AI race is moving out of the concept phase and deeper into the industrial phase. The companies that win from here will not just be software brands and GPU giants. They will include the firms solving the uglier, less glamorous problems inside AI infrastructure: power conversion, electrical efficiency, rack density, and grid stress. |
That is why Navitas Semiconductor is interesting under $20. |
It is small.
It is speculative.
Its current numbers are weak.
And it is absolutely not a low-risk idea. |
But if Oracle's quarter is telling the truth about where AI infrastructure is headed, then the market will eventually spend a lot more time on the companies that make high-density power delivery possible. |
Navitas is one of the few sub-$20 names with a credible shot at belonging in that conversation. |
Disclaimer: This article is for informational purposes only and does not constitute investment advice. Investing involves risk, including the potential loss of principal. Always do your own research before making investment decisions. |
0 التعليقات:
إرسال تعليق