Gene Munster

Jensen reminds investors of the big picture: Over the past five years, they have moved from a chip company to an AI factory company. He says there is a lot of compute that is squandered today. Says systems are getting more complex, and Nvidia is the platform to build the most cost-effective infrastructure because they bring together CUDA, compute, and networking. My take: Factoring in Jensen's demand comments for CY27, the Street will likely increase revenue growth estimates from 30% to 39%. For CY28, the Street is at 16% growth. I don’t expect analysts will change that number after today. I believe CY28 will finish measurably higher than t…

Jensen brings in the Groq acquisition related to higher priced tokens. Says the economics of generating tokens is not a race to the bottom. He argues it’s a race to the top, the smartest models users will pay the most for. Token pricing will be increasingly segmented into tiers. Tier 1: High throughput and low latency (often free or low-cost). Tier 2: Larger models, specialized pricing. Tier "N": The largest, smartest models. The more input token context, the smarter the models, the price goes up. This will be Nvidia's most advanced GPU ,Groq 3 into the Vera Rubin platform My take: Makes sense. Not all AI is created equal, and premium t…

Jensen reminds investors that the cost of $NVDA compute is still the lowest when you consider price and output. My take: True and investor are still waiting for the day when that changes. Part of the wall of worry.

The fact that Jensen guided CY27 revenue about 7% ahead of the Street, yet the stock only moved about 1% higher on the news, speaks to investors' willingness to believe. $NVDA

Big update on demand: Stock goes from up 2% to up 3%. Raised CY27 revenue to be about 7% higher than the Street. Specifics: Jensen now says through end of CY27 there is $1 trillion in demand for Rubin. Three weeks ago, he was saying CY26 was going to be better than $500B. My take: This implies CY27 revenue of better than $500B. The Street is currently at $468B, so he effectively raised next year’s outlook 7% higher than the Street. $NVDA

Stock check in: $NVDA is still up 2% today, vs. up 2.8% at the start of the keynote. Nasdaq up 1.5%.

$NVDA keynote kicks off any minute. Its a tie for key topics today: 1. Rubin Demand 2. How they'll benefit from inference. My take: 1. Jensen will likely say Rubin demand continues to be off the charts, and that's what investors already know. Unfortunately a non event for the stock. 2. Inference topic is one of the keys for the next five years. I'm all ears.

Poll: What’s the key topic for Jensen’s keynote today at 2p ET? $NVDA

Getting ready for Jensen’s keynote on Monday. His message to a skeptical investing crowd: we’re at an AI utility inflection point. $NVDA https://t.co/8yQIukz8NN

🚀

Reflecting on Apple’s upcoming 50th birthday reveals why the company will continue to be successful. $AAPL https://t.co/iNP8uYANX9

The “yin and yang” of the future of software. On the headcount reduction yin: • March 11: Atlassian reduces headcount by 10%. “It would be disingenuous to pretend AI doesn’t change the mix of skills we need or the number of roles required in certain areas. It does,” CEO Mike Cannon-Brookes. • Feb 26: Block reduces headcount by 40% (not entirely because of AI): “Intelligence tools are making a new way of working.” • Jan 27: Pinterest lays off 15% to “focus more on AI products.” On the stronger need for AI yang: • Atlassian reported 26% y/y cloud revenue growth (for Q2 of its 2026 fiscal year). Bottom line: Software is being rebuilt for …

“SaaSpocalypse” update: Jensen Huang noted at the Morgan Stanley TMT Conference that software will look far different in the future than it does today. He argued the, “software industry in the future will be much larger,” as companies evolve from classic seat-based models toward renting agents and specialized tokens. My take: Seat-based models will face pressure as the knowledge workforce contracts. Renting agents, or usage-based software companies, will flourish. We believe $LOUP offers diversified exposure to the agentic software cycle Jensen is describing. To learn more about the Innovator Deepwater Frontier Tech ETF ($LOUP), visit: http…

Next year's big tech capex will be meaningfully higher than where the Street is at in part because of today’s Ratepayer Protection Pledge from the White House. ➡️Tech is happy to pay for power. ➡️This marks a shift in the AI Arms Race by laying from silicon to power. ➡️This will increase our reliance on Google, Meta, Amazon and Meta. ➡️The cost will translate to higher capex for longer. 2026, the mega caps will increase capex by about 65% compared to 70% last year. ➡️2027, the Street is expect mega cap capex to increase by 14%. I expect it to be 40%. ➡️Currently the Street is at 14%. The numbers: Google 13%, Amazon 6%, Meta 13%, Microsoft 19…

The new $AAPL MacBook Neo should add 0.5% to overall revenue for the next year. I believe it won't impact margins. More on Pressure Points: https://t.co/EPLseHazXH

The new Mac’s announced today add color to Cook’s “range of options to deal with margins”. $AAPL https://t.co/8LuTWFqb49

The new $AAPL products should translate into a modest upside for June quarter revenue. https://t.co/3YfhPcJTiW

This is a special week for $AAPL with the first ever three day drip of new products. So far, the 17e and iPad Air specs and pricing are what we expected and translates to a small positive for revenue over the next couple of quarters. I'm looking out for that lower priced Mac.

Speaking of keeping the same $599 entry price, the new iPad Air announced today doesn't move the needle. iPad accounts for 7% of revenue. The Air accounts for about half of that, or 3.5% of sales. A modest short-term bump is largely irrelevant. Shares of $AAPL are down 0.25%, in line with the Nasdaq.

$AAPL announces (as expected) iPhone 17e. Same $599 starting price as the 16e, but with 256GB of storage vs. 128GB on the 16e and better chip. My take: The impact on iPhone sales will be a around a 2% positive for the next couple of quarters (math below). Since Apple already had a $599 option, any uplift should come from price-focused buyers who were waiting for the new version. The math: I estimate the "e" models account for about 20% of units in March and June, then decline to 15% of units. That means if the iPhone 17e increases growth in that segment by 15% for a couple of quarters, then normalizes, that would increase overall units by 3…

Two days after strong $NVDA earnings, the stock continues to slide, which begs the question: Will fundamentals ever be enough? https://t.co/IJF2dA07mR

AI’s impact just hit a new gear. Block ($XYZ) is cutting 40% of its 10k-person headcount due to AI efficiencies. My take: we’re in the second inning.

$NVDA Three takeaways on Pressure Points: 1. The stock being down 1% underscores that investors’ long term growth concerns remain. 2. The fundamentals and outlook came in better than elevated whisper numbers. 3. Most importantly, the guidance reinforces just how early we are in AI.

The final question of the call turned into a 7-minute Jensen manifesto on how growth can continue. $NVDA My take: It passed the sniff test. Here’s the play-by-play: Goldman asked if Jensen still feels good about growth accelerating beyond the April quarter. Jensen took a "ground-up" approach to the answer: The Scale: The world spends $300B–$400B on classical computing, but AI requires 1,000x more. If the world sees value in the AI "token," it will invest to produce it. The Spend: Infrastructure spend to meet token demand will exceed $700B this year driven by enterprise software, robotic factories, agent-based small businesses, and cars t…

Fun question about space data centers: Jensen says the economics are poor today but will improve over time. He adds space offers an abundance of energy and real estate, but cooling is the bottleneck. Addresses the cooling concern that there’s no air for convection, and you need massive radiator which is a total departure from Earth-bound methods. Hopper is already in space, and "imaging" is a key use case. My take: Jensen message is orbital data centers are difficult today and worth pursuing. $NVDA

$NVDA is now up only 0.6%, compared to 2.5% when the call began. One reason for the drift: the call is missing the usual "early innings" table pounding message. Jensen is leading with data over his typical bombardment of bullish adjectives. My take: We’re still in the 2nd inning.

Jensen says we’re at the inflection point where compute equals revenue. He points to the ROI from Claude Cowork, OpenClaw, and Claude Code as proof that AI is shifting from an expense to a direct profit driver. My take: Strong narrative, but it won’t silence the skeptics who fear AI will result in profitless prosperity, massive productivity gains that don't have a ROI.

Jensen takes the mic for three minutes: OpenAI: Confirmed they are nearing an expanded partnership (following reports of a $30B+ equity stake). Meta: Says things are "going well," though avoided addressing Meta’s recent pivot to AMD for certain workloads. Then turned it over to Q&A.

Kress says they’ve secured inventory to meet Data Center demand for the next several quarters. My take: The stock dipped by 1% on the news. Ironically, operating outside of supply-demand equilibrium is often better for the stock; it allows investors to "dream" of what results would have been without constraints. $NVDA

Physical AI was about 3% of revenue in CY25. CFO says this will be a major driver. My take: I believe by CY30 this segment can become a $50B plus annual business, up from about $6B in CY25. That implies physical AI could represent about 8% of a $600B revenue base in CY30. $NVDA