Is Nvidia currently the "most important stock in the world"?

For Nvidia investors, the past two years have been a joyful ride. However, recently, they have experienced more ups and downs. As the main beneficiary of the artificial intelligence (AI) boom, Nvidia’s market capitalization has surged nearly ninefold since the end of 2022. However, after reaching a record high in June and briefly becoming the most valuable public company in the world, Nvidia then lost nearly 30% of its value over the next seven weeks, shedding around $800 billion in market capitalization.

Now, the stock has rebounded significantly, sitting just 6% below its all-time high.

Can Nvidia shape AI expectations?

With the chipmaker set to release its quarterly earnings on August 28, the stock's volatility is a primary concern on Wall Street. Any sign that AI demand is slowing or that a top cloud computing customer is modestly tightening its spending could lead to a significant decline in future revenues.

Eric Jackson of EMJ Capital told CNBC's "Closing Bell" last week: "This is the most important stock in the world right now. If they don’t perform well, it will be a big problem for the entire market. I think they will deliver positive surprises.”

Nvidia’s report comes a few weeks after other tech giants released their earnings reports. The name “Nvidia” echoed throughout those analyst calls, as Microsoft, Alphabet, Meta, Amazon, and Tesla all have been heavily investing in Nvidia’s graphics processing units (GPUs) to train AI models and run large workloads.

In Nvidia's past three quarters, revenue has more than tripled year-over

Nvidia CEO Jensen Huang

Much of the optimism ahead of the report—shares have surged nearly 10% in August—is driven by reports that top customers will continue to heavily invest in data centers and Nvidia-based infrastructure. On August 27th, shares rose 1.5%, closing at $128.30.

Last month, the CEOs of Google and Meta enthusiastically endorsed the pace of their build-outs, stating that underinvestment poses a greater risk than overspending. Former Google CEO Eric Schmidt recently told Stanford students that he’s hearing from leading tech companies that they need processors worth $20 billion, $50 billion, or even $100 billion.

However, despite Nvidia’s margins improving recently, the company still faces questions about the long-term return on investment that customers will see from purchasing devices priced at tens of thousands of dollars each, being ordered in large quantities.

In Nvidia’s most recent earnings call in May, CFO Colette Kress provided data points suggesting that cloud providers, which account for over 40% of Nvidia’s revenue, will generate $5 in revenue for every $1 spent on Nvidia chips over four years.

Numerous statistics like these are expected to be released. Last month, Goldman Sachs analysts wrote, following a meeting with Nvidia’s CFO Colette Kress, that the company would share additional ROI metrics this quarter "to instill confidence among investors."

Timing of the Blackwell AI Chip Launch

Another major question for Nvidia is the timing of the launch of its next-generation AI chip, named Blackwell. Earlier this month, The Information reported that the company is facing production issues, which could delay large shipments to the first quarter of 2025.

This news comes after Nvidia CEO Jensen Huang surprised investors and analysts in May by stating that the company would recognize "a lot" of revenue from Blackwell in the current fiscal year.

Nvidia’s current generation of chips, known as Hopper, remains the top choice for deploying AI applications like ChatGPT. However, with emerging competition from Advanced Micro Devices, Google, and a host of startups, Nvidia must do everything it can to maintain its leadership in performance through new chip releases.

Even with the potential delay of Blackwell, the revenue from Blackwell chips might only be pushed back by about a quarter. Meanwhile, sales of the current Hopper chips, particularly the newer Hopper H200, will be bolstered.

Morgan Stanley analysts wrote in a report this week: "The shift in launch timing isn’t very critical, as both supply and customer demand have quickly pivoted to the H200."

Many of Nvidia’s top customers say they need the additional processing power of Blackwell chips to train more advanced next-generation AI models.

Frank Lee, an analyst at HSBC, wrote in an August report: "We expect Nvidia to scale back the allocation of Blackwell B100/B200 GPUs to prioritize ramping up production of the H200 Hopper in the second half of the year."