Nvidia shares fell nearly 23% from mid-July through early August, closing on August 7 at $98.91. The selloff was triggered by fears that Nvidia’s big tech customers may trim their spending budgets, as well as rumored delays in the rollout of Nvidia’s Blackwell GPUs slated for launch later this year. The stock drawdown was further exacerbated by some high profile investment funds exiting or cutting stake in Nvidia, and doomsday predictions Nvidia’s shares were running rampant on Wall Street. However, the stock staged a strong recovery of more than 30% from these low levels, closing August 19 at $130. Even as the AI bellwether braces to release its second-quarter earnings report on August 28, what should investors expect from Nvidia’s upcoming earnings print? Is it time to sell Nvidia after its nearly 3,000% rally in the past five years?
When is a good time to sell or any other stock? When the stock fundamentals have deteriorated and do not support the current price levels at which the stock trades. Is that true of Nvidia?
After a meteoric rise in the past years, Nvidia stock lost its sheen somewhat after its recent slump of more than 20%. The selloff was triggered by concerns related to a possible capex slowdown by Nvidia’s top customers and hyperscalers Microsoft (MSFT), Meta Platforms (META), Amazon (AMZN) and .
However, all these companies have pledged to boost their spending on even higher. Google parent Alphabet said its future capex will be “at or above the first-quarter level of $12 billion.”
Microsoft, which is Nvidia’s largest customer, reported capital spending of $55.7 billion for the fiscal year (ended in June) up 75% from last year, and noted that capital spending in fiscal 2025 will top that level to meet the growing demand signal for its and cloud products.
Meta stated that it currently expects significant capex growth in 2025 as it invests to support its AI research and product development efforts. The elevated AI capital spending appears to be paying off for Meta, which raised its third-quarter revenue guidance to $38.5 billion to $41 billion vs. the $39 billion consensus estimate, thanks to a robust growth in advertising. Meta has been leveraging AI to optimize ad targeting, which has significantly improved its advertising performance and business efficiencies. Amazon expects capital investments to be higher in the second half of the year and the majority of the spend will be to support the growing need for AWS infrastructure as the company continues to see strong demand in both generative AI and non-generative AI workloads. JPMorgan sees the four hyperscalers–Google, Microsoft, Meta and Amazon–spending a collective $200 billion in capex this year with AI buildout dominating the capital spending, which in turn bodes well for Nvidia.
Another cause for concern centers around the rumored delay in the rollout of Nvidia’s next generation Blackwell AI chips, which are roughly two times faster than Nvidia’s current Hopper models, but with notable improvements in energy efficiency. The delay is reportedly attributed to the complexity of the chip-on-wafer-on-substrate (CoWoS) packaging technology used by Taiwan Semiconductor Manufacturing Company (TSM) that manufactures these chips for Nvidia. The Blackwell launch is scheduled for later this year.
However, Nvidia commented that Blackwell sampling has started, and production is on track to ramp in the second half. Nvidia’s hardware partners, like Foxconn, Quanta, Wistron, Pegatron, and Asus, have already demonstrated their Blackwell-based servers at Computex, a computer expo held annually in Taipei, Taiwan. Investors are hoping that Nvidia may provide a more specific timeline for the Blackwell ramp in its upcoming earnings call, beyond an ambiguous reference to the second half of the year.
Even assuming there is a delay in a worst case scenario, it is good to know that Nvidia is working on fixing the design complexities rather than bringing out a flawed model that is bound to fail. The delay (if there is one) is unlikely to hamper Nvidia or its prospects in the longer run. Don’t forget, Nvidia is still the undisputed leader in the data center GPU market, and even if AMD brings out a rival product comparable to Blackwell, it will still have to catch up with the yawning chasm in market share.
Besides, there’s the alluring revenue dynamics at play. According to news reports that reference a Morgan Stanley analysis, Nvidia and its partners are expected to price an AI server cabinet, featuring the upcoming Blackwell GPUs, between $2 million and $3 millionMumbai Stock Exchange. This pricing could potentially lead to an estimated annual revenue exceeding $200 billion in 2025 based on the requirement for tens of thousands of AI servers. If were to be trusted, Nvidia may have boosted its Blackwell orders with TSMC by 25% and that could be favorable for TSMC, which is manufacturing the most powerful AI chip in the world based on Blackwell architecture.
The much-touted exodus of top money managers from Nvidia shares during the second quarter is reflected in some 13F filings. The SEC’s Form 13F must be filed every quarter by institutional investment managers with at least $100 million in assets under management (AUM).
According to the latest 13F filings,
Stanley Druckenmiller’s Duquesne Family Office slashed its stake in Nvidia by 88% and now holds less than 1% of Nvidia in its portfolio.
David Tepper’s Appaloosa Management reduced its Nvidia holdings by 84.4% and now Nvidia is just 1.38% of its portfolio.
Soros Capital Management LLC, Twin Tree Management LP and Paul Singer’s Elliott Investment Management Holdings exited their positions in Nvidia.
Light Street Capital Management Holdings cut its Nvidia stake by 26.1% but Nvidia is still its top holding representing 17.1% of its portfolio.
Lee Ainslie’s hedge fund Maverick Capital trimmed its stake in Nvidia by 2.86% but Nvidia is still among its top five holdings representing 4.8% of its portfolio.
Whale Rock Capital Management Holdings cut its stake by 40.7% in Nvidia, which still remains its top holding representing 8.5% of its portfolio.
Interestingly, the top shareholders of Nvidia shares– and –added 11.4 million and 16.9 million shares, respectively during the second quarter.
While it may seem like a great strategy to follow “smart money,” it should be noted that the 13F is just one among several data points for investors to make informed investment decisions. Besides, the 13-F can be filed up to 45 days after the end of a quarter, meaning that the filing only gives a peek into past strategies of institutional investment managers and may not hold much relevance at the time when it comes into public knowledge. Also, institutional investment managers have deep pockets, and specific investment strategies which they are not required to reveal to the public fully, like their short positions for example. So attempting to reverse-engineer their success without understanding the underlying strategy can often burn the investor.
At this point, does it look like there is any compelling reason to sell Nvidia shares?
As a key enabler of AI with its GPUs powering even supercomputers used by Meta and , Nvidia continues to dominate the AI chip market with an estimated market share ranging from 70% to 95%. For a hardware company, Nvidia’s gross margins are quite high at 73.8% for fiscal 2024. Nvidia’s competitive moat lies in CUDA, its proprietary software stack that allows developers to leverage the parallel processing capabilities of Nvidia GPUs to accelerate machine learning workloads. Attempts to migrate from the CUDA have gone south, but Nvidia is never negligent. It constantly evolves CUDA’s capabilities, to retain its market leadership. But that hasn’t deterred AMD, Intel and Google from persevering in valiant attempts to unseat the CUDA from its pedestal.
Nvidia’s upcoming Blackwell platform has an inference capability that is 30x of the incumbent Hopper’s, while consuming 25x less cost and energy. Nvidia has dismissed any concerns of customers holding off on Hopper orders because of the upcoming Blackwell launch. The demand for both Hopper and Blackwell platforms is well ahead of supply and this is expected to continue well into the next year. The upcoming earnings report will provide more clues on the Blackwell ramp.
Nvidia sees a long-term market opportunity of $1 trillion with $300 billion from datacenter, $100 billion from gaming, $300 billion from autonomous vehicles and robotics, $150 billion from industrial meta verse via its Omniverse ecosystem and $150 billion from Enterprise via Nvidia AI Enterprise–its operating system for enterprise AI, as well as DGX Cloud that gives customers instant access to Nvidia AI supercomputing in global-scale clouds.
Nvidia’s growth has been strong and profitable. In the last five years, revenue has grown from $10.9 billion in fiscal 2020 to $60.9 billion in fiscal 2024 with revenue reaching $26 billion for the first quarter of 2025, while operating margins have risen from 34% to 61%, and at 69% for the first quarter.
Data Center revenues constitute a major chunk of its top line, and have grown at a 75% CAGR in the past five years to $47.5 billion in fiscal 2024. In the first quarter, data center revenues were $22.6 billion.
Gaming revenue, which represented 17% of Nvidia’s fiscal 2024 top line, has grown at a five-year CAGR of 11% from $5.5 billion in fiscal 2020 to $10.4 billion in fiscal 2024.
Nvidia’s Professional Visualization segment represented 3% of 2024 revenues, but is seen as a significant future growth driver. The business offers Omniverse as a development platform for enhancing productivity and introducing new capabilities in design, manufacturing and digital content creation.
Nvidia’s automotive revenues stem from its platform solutions for automated driving and in-vehicle cockpit computing. The Nvidia DRIVE
is an end-to-end Autonomous Vehicle (AV) platform with a full software stack powered by systems-on-a-chip (SoCs) in the vehicleBangalore Stock Exchange. Xiaomi’s first electric vehicle, the SU7 sedan is built on the Nvidia Drive Orin, which is Nvidia’s AI car computer for software-defined autonomous vehicle fleets. Nvidia Drive Thor, the successor to Orin, has already procured design wins with EV makers, and slated for production in vehicles in 2025. Although automotive revenues represented only 2% of the top line in fiscal 2024, this is another key area of growth for Nvidia in the future, amid the rising momentum of self-driving vehicles. The global autonomous vehicle market is to grow to $448.6 billion by 2035, according to Allied Market Research.
Financial Health Indicators (H2)Nvidia’s free cash flow (FCF) has grown from $4.3 billion in fiscal 2020 to an impressive $26.9 billion in fiscal 2024, with first quarter FCF at a stellar $14.9 billion. In fiscal 2024, the company spent $395 million in dividends, which it plans to maintain, and $9.5 billion on stock buybacks, reflecting its strong alignment with shareholders. It is not surprising considering that Nvidia’s Founder CEO Jensen Huang owns more than a 3% stake in Nvidia. There was some controversy about the CEO offloading some shares and making good profits before Nvidia shares dropped. Although the timing of the sale turned out to be unusual, the shares were sold under a 10b5-1 trading arrangement that allows company insiders to sell company stock on a predetermined schedule without breaching insider trading laws. So, there’s no cause for concern. Like other shareholders, Jensen Huang experienced a significant paper loss when Nvidia’s stock value fell in the recent period.
AI is a secular , not a one-off pattern to fizzle out. According to management consulting firm MarketsandMarkets, AI is estimated to grow from a value of $214.6 billion in 2024 to $1,339.1 billion by 2030, and Nvidia’s strong positioning makes it a key beneficiary of these AI tailwinds. However, Nvidia’s growth is expected to normalize from the fast, furious and feverish pace to stable and more sustainable levels.
A low interest rate environment is typically better for tech companies, as it spurs R&D spending, deal making (M&A) and fundraising in the tech sector. However, AI has been the winning theme so far, defying the higher-for-longer interest environment, by shaping capital spending plans of large tech enterprises.
Capex of mega cap techs are at new highs, and the budget is expected to expand further in 2025. Tech spending intent by corporations remained positive in the second quarter of 2024, as reflected by the U.S. Technology Demand Indicator () score of 51.71 in the second quarter (as measured by 451 Research). The score is slightly lower than the first-quarter number of 52.11, but still positive as a score above 50 typically indicates expansion. This bodes well for Nvidia, which is a key beneficiary of AI tech spending. With the wide expectation of Fed interest rate cuts starting in September, the capital spending environment in the tech sector is likely to get even more conducive.
The global gaming market is estimated to generate $187.7 billion in revenue in 2024 with nearly half of it originating from mobile games, according to gaming market data and research firm . Nvidia’s flagship gaming GPU product line GeForce, supports more than 200 million gamers. The latest RTX 40-series graphics cards of GeForce based on Ada Lovelace architecture continue to stay on top of the game. The gaming industry is looking forward to Nvidia’s next-generation GeForce RTX 50-series graphics cards based on the Blackwell architecture. Significant growth in e-sports–the gaming competitions for professional players and teams–also highlights the demand for high speed and performance packed Nvidia GPUs.
Industrial Metaverse is a virtual environment that will apply metaverse elements like digital twins, virtual reality (VR) and augmented reality (AR
) to industrial applications, to enhance the efficiencies of industrial operations and processes. A simple example would be the maintenance team in a manufacturing company applying VR to simulate repairs on a digital twin of a machine and optimizing the procedure before actually applying it to a physical machineChennai Investment. Allowing businesses to model prototypes and test in a digital environment before committing physical and human resources to a project, will enhance operational efficiency, reduce downtime, and improve overall productivity, while unlocking immense value for enterprises. Nvidia’s Omniverse is a scalable, multi-GPU real-time development platform for building and operating metaverse applications. Nvidia is collaborating with Siemens to build the industrial Metaverse and to increase use of AI-driven digital twin technology for enhancing industrial automation. estimates the global industrial metaverse market to reach $228.6 billion by 2029 from an estimated $28.7 billion in 2024, with an estimated CAGR of 51.5%, thanks to the rising adoption of digital twins, advancements in AR, VR, AI, IoT, and rising demand for efficiency and optimization in the industrial sector. As a key player, Nvidia is well positioned to benefit from this growth.
A citing Jeffries analysts noted that the U.S. may implement new trade restrictions that could ban Nvidia from selling its H20 AI chips to China. H20 is Nvidia’s specially designed chip for China, after regulators tightened restrictions on selling high-end AI chips to China, including Nvidia’s H100, citing national security concerns. However, H20’s computing power is significantly lesser compared to H100, to achieve compliance with the U.S. sanctions. But, now even the H20 may possibly be banned for sale in China when the U.S. reviews its semiconductor export controls in October, says the report and if the ban occurs, Nvidia stands to lose an estimated in revenue. Analysts had previously projected that Nvidia will deliver more than 1 million new H20 chips to China, and generate more than $12 billion in sales as each H20 chip is priced between $12,000 and $13,000. In the prior financial year, Nvidia’s China revenues were estimated at $10.3 billion.
According to a report, Huawei is preparing to challenge Nvidia’s H100 AI chips with its Ascend 910C, and expects to begin shipping these chips as early as October. The Ascend 910C chip is reportedly being tested by Chinese companies, while TikTok parent ByteDance, Baidu and China Mobile may be engaged in early discussions to buy the chip. If the news reports were to be believed, this could pose a serious threat to Nvidia’s market share in China, which is already very limited because of U.S. trade restrictions.
Nvidia is scheduled to release second-quarter 2025 earnings on August 28. Here’s a rundown of what investors can expect…
Analysts expect Nvidia to report second quarter adjusted earnings of 64 cents a share on revenues of $28.5 billion, representing a more than two-fold increase from year-ago’s split-adjusted 27 cents per share on $13.51 billion revenues.
Nvidia appears well positioned to beat earnings and revenue estimates for the second quarter, if we look to the last four quarters for direction. Nvidia beat EPS and revenue estimates in the last four quarters, with the magnitude of beats reducing sequentially. (See table below.)
A definite time frame for the Blackwell ramp would be helpful, if Nvidia wants to put the delay rumors to rest. Nvidia had said it expects to see a lot of Blackwell revenues this year. Analysts project Blackwell revenues exceeding $200 billion for Nvidia next year.
Nvidia is well positioned to benefit from its top customers’ plans to continue spending heavily on AI-related infrastructure.
Fundamentally, nothing has changed to impact Nvidia’s competitive edge in AI or accelerated computing. Among peers, Nvidia is still an enviable market leader in AI chips. The demand for both Hopper and Blackwell platforms is well ahead of supply and is expected to continue well into the next year. Strong and accelerating demand for generative AI training and inference should propel data center growth.
Nvidia may provide more color on its new Spectrum-X Ethernet networking solution that began shipping in the first quarter. The Spectrum-X enables Ethernet-only data centers to accommodate large-scale AI. Nvidia expects the Spectrum-X to ramp to a multibillion-dollar product line within a year.
An update can be expected on Sovereign AI, which refers to a nation’s capabilities to produce AI using its own infrastructure, data, workforce and business networks to navigate the complex landscape of data privacy and security. Nvidia sees its Sovereign AI revenue approaching the high single-digit billions this year, from nothing last year.
An update may be expected on Omniverse industrial digitalization, which is expected to drive the next wave of growth in the professional visualization business.
Nvidia sees its automotive business driving a multibillion revenue opportunity across on-prem and cloud consumption. The second quarter earnings call may provide more clues on the demand for the Nvidia Drive Thor and the Nvidia Drive Orin.
Any update on Nvidia’s China sales/market will be a key watch point.
In the past five quarters, Nvidia shares have continued to set new highs in the days and weeks following a strong earnings report and guidance. However, after Nvidia reported spectacular results for the second quarter of 2023, the stock rose by a lackluster 6% although that was still a new high. Was it because the Nvidia stock had run up significantly ahead of its second quarter announcement? Will that pattern repeat this year?
Source: Yahoo Finance
Nvidia is firing on all cylinders. Fundamentally nothing has changed much for Nvidia. It is still the market leader for AI chips and demand for its GPUs appears as strong as ever with its top customers having pledged to keep spending heavily on AI infrastructure. Any delay in the Blackwell ramp may cause near-term volatilities, but the true risk lies in the likelihood of a trade ban that would stop it from selling its H20 AI chips to China, the second largest economy in the world. But if the company loses an estimated $12 billion in China revenues, there is still the $200 billion in estimated revenue potential for Blackwell. The true competitive edge of Nvidia lies in the strong moat it builds around its products, like the CUDA for instance.
Nvidia could report a stellar second quarter, but will the Nvidia stock rally in the near-term? That depends. If the stock runs up ahead of its second-quarter earnings, then the post-earnings rally may likely be lackadaisical. But, Nvidia stock is for the long haul, as AI is here to stay. The strategy would be to buy on dips.
Please note that I am not a registered investment advisor and readers should do their own due diligence before investing in this or any other stock. I am not responsible for the investment decisions made by individuals after reading this article. Readers are asked not to rely on the opinions and analysis expressed in the article and encouraged to do their own research before investing.
Jaipur Stock