Welcome back, everyone.
Hope you all had a great session, as the levels once again worked out in our favor. We called the 6920 long level with a target of 6960, which hit just below the high of the day. As a result, the entire rally was captured between our long level and our target.
We then expected sellers to step in at the end of the day to drive a move down of 10 to 20 handles. Unfortunately, this fell short; while contracts saw some upside, it wasn’t to my liking since the targets were lower. Overall, it was a great session, and we can now look toward the earnings reported after the close.
Nvidia Earnings
The fiscal Q4 2026 earnings report from NVIDIA serves as a definitive marker for the current state of the global technology landscape, delivering a performance that exceeded even the most aggressive Wall Street estimates. The company posted record-breaking results across every critical metric, including total revenue, data center sales, earnings per share, and gross margins. While these numbers provide the quantitative proof of NVIDIA’s dominance, the true narrative of the report lies in CEO Jensen Huang’s forward-looking commentary. Huang signaled that the industry has moved beyond a simple transition to AI and has instead hit a massive inflection point centered on agentic AI. This new era shifts the focus from models that merely answer questions to autonomous systems capable of executing complex workflows and tasks. According to Huang, this shift has driven AI compute demand to levels over 1,000 times greater than that of traditional general-purpose computing.
The transition to the Blackwell architecture is a cornerstone of this growth, with demand strengthening significantly as the industry balances the massive compute needs of training new models with the rapidly expanding requirements of running them, known as inference. A particularly striking development highlighted during the call was the rise of Sovereign AI. Nations across the globe are increasingly treating AI infrastructure as a critical national resource, similar to energy or telecommunications. NVIDIA projects that as global GDP grows, these national investments will provide a sustained floor for demand that is independent of the typical Silicon Valley venture cycle. Perhaps the most confident signal sent to the market was NVIDIA’s guidance regarding China; despite anticipating zero data center compute revenue from the region due to trade restrictions, the company still expects to deliver record-breaking overall growth, effectively demonstrating that the rest of the world’s appetite for their chips is more than enough to offset the loss of a major geographic market.
Looking toward the remainder of 2026 and into 2027, NVIDIA’s management team expressed an almost defiant level of confidence in their trajectory. They suggested that previous revenue projections for the Blackwell and upcoming Rubin product lines might actually be conservative, given the sheer volume of the current order backlog. To ensure they can meet this relentless demand, the company has already made multi-billion dollar supply chain commitments to secure manufacturing capacity through 2027. This aggressive positioning is partly a response to recent skepticism in the software sector. Huang addressed the recent sell-off in software stocks directly, calling the market’s reaction “out of touch with reality.” He argued that the market is unfairly punishing these companies for temporary earnings fluctuations, failing to see that AI agents will eventually become the primary engine for software revenue. In Huang’s view, the software “bluff” will eventually be called as these platforms begin to show the true productivity gains promised by the hardware layer.
However, despite this corporate optimism, a tension remains between NVIDIA’s parabolic growth and the broader market’s ability to digest it. There is a fine line between setting ambitious projections and maintaining the trust of investors when those targets are tested by reality. As seen with companies like Tesla, a visionary leader and a massive head start in a new industry can sustain a stock rally even when certain milestones are missed, provided the underlying belief in the mission remains intact. For NVIDIA, the mission is the total replacement of traditional data centers with AI factories. The risk, as noted by many observers, is the potential for a plateau in AI capabilities. While the initial leap in AI performance was eye-catching and transformative, there is a growing concern that subsequent iterations may offer diminishing returns. If AI development hits a wall where new models are no longer “pushing the envelope” in a way that justifies tens of billions in new spending, NVIDIA may eventually face the same gravity that affects all infrastructure cycles. For now, however, NVIDIA remains the indispensable backbone of the movement, operating under the mantra that without their silicon, there simply is no AI.
To understand if NVIDIA’s bold projections are anchored in reality, one must look at the capital expenditure (CapEx) plans of their largest customers—Microsoft, Alphabet, Amazon, and Meta—who collectively account for the vast majority of high-end GPU demand. The data from the early 2026 earnings season suggests that NVIDIA’s confidence is backed by a historic and accelerating surge in spending. These four tech giants have signaled a collective 2026 investment package of approximately $650 billion, a staggering 60% increase from the already record-breaking $410 billion spent in 2025. This massive pool of capital is being funneled directly into the data centers, silicon, and power infrastructure required to run the very Blackwell and Rubin systems Jensen Huang highlighted in his call.
Microsoft has positioned itself at the forefront of this industrial shift, reporting a record $37.5 billion in capital expenditures for a single quarter alone. Roughly two-thirds of this spend is dedicated to short-lived assets like GPUs and CPUs, effectively serving as a direct revenue pipeline to NVIDIA. Furthermore, Microsoft’s commercial remaining performance obligation (RPO) has surged to $625 billion, with nearly half of that balance tied to its partnership with OpenAI. This indicates a massive, multi-year backlog of demand for AI compute that is already legally committed, providing strong evidence that NVIDIA’s supply commitments through 2027 are not merely speculative but are indexed to signed enterprise contracts.
Alphabet and Amazon have mirrored this aggressive posture, with Alphabet raising its 2026 CapEx guidance to a range of $175 billion to $185 billion and Amazon committing to a $200 billion investment plan. While investors have expressed some nervousness regarding the near-term return on investment (ROI)—leading to temporary pullbacks in their respective stock prices—the companies themselves are doubling down. Amazon’s leadership specifically noted that the cost of “not investing” in AI infrastructure is higher than the risk of overspending, as they race to ensure they have the “inference power” necessary to host the next generation of agentic AI applications. This “fear of missing out” at a corporate scale creates a secular tailwind for NVIDIA that is largely insulated from minor fluctuations in the consumer software market.
Meta’s strategy further solidifies the long-term outlook for NVIDIA’s newer architectures. Mark Zuckerberg’s team recently raised their 2026 CapEx outlook to a range of $115 billion to $135 billion, specifically citing plans to deploy “millions” of NVIDIA Blackwell and Rubin GPUs. Meta is not just using these for internal research but is industrializing its advertising engine with AI, transforming its social media platforms into high-performance compute environments. By securing long-lead inputs like gigawatt-scale power agreements and multi-year fiber contracts, Meta is effectively building a “moat” of physical infrastructure. When viewed in aggregate, the sheer scale of these commitments from the world’s largest companies suggests that while the “capabilities” of AI may eventually face a hurdle, the physical build-out of the “AI factories” is locked in for the next several years.
Despite this technical and financial dominance, the sustainability of this “AI gold rush” is increasingly being questioned through the lens of infrastructure and return on investment. The most immediate physical bottleneck is the global power grid. The extreme energy density required by NVIDIA’s latest chips has forced hyperscalers to transition from being software companies to becoming energy developers, investing heavily in private natural gas and nuclear solutions to bypass aging public grids. If the physical ability to plug in these chips cannot keep pace with the manufacturing of the chips themselves, the industry could face a hard ceiling. Furthermore, there is a growing “ROI gap” where the trillions being spent on hardware must eventually be justified by equivalent revenue from AI software. While Jensen Huang dismissed recent software stock sell-offs as being “out of touch with reality,” the market remains wary of whether enterprise customers will see the productivity gains necessary to fund this level of capital expenditure indefinitely.
Despite the overwhelming financial performance and the aggressive capital expenditure from its largest customers, a notable tension exists in the market: NVIDIA’s stock has yet to break through to new all-time highs following this report. This divergence between record-breaking fundamentals and stagnant price action suggests that the market may have already “priced in” much of this perfection. While the company is delivering on every internal metric, investors appear to be grappling with the law of large numbers. There is a growing sentiment that for a company already valued in the trillions, a “beat and raise” is no longer a catalyst for a rally but rather the minimum requirement to prevent a sell-off. This suggests that the bar for a true breakout has moved from simply reporting strong numbers to proving the long-term utility and profitability of the software layer built upon this hardware.
This stagnation also reflects a broader cautiousness regarding the sustainability of the “AI premium.” Even with the massive influx of capital from hyperscalers, the fact that the stock is hovering below previous peaks indicates that some investors are waiting for tangible evidence of the “inference” revenue Jensen Huang described. There is a psychological barrier at play; as the company’s valuation reaches unprecedented levels, the market demands more than just future projections—it seeks proof that the massive build-out of AI factories will not lead to a glut of unused compute capacity. Until there is a clear sign that the end-users of AI are generating the cash flow necessary to sustain this cycle through 2027 and beyond, the stock may continue to consolidate despite the record-breaking numbers.
There is a sea of other stocks to participate in; as of right now, I think Nvidia is a pass for me. Everything is currently priced in. This stock could very well sit and do nothing for months to come, and I believe there are better investment opportunities out there. If you are holding long-term, you will likely do great, but in the near-term, the stock needs time to cool off.
Now that the main earnings are out of the way, we can focus on what this all means for the S&P.
