With Nvidia's highly anticipated fourth-quarter earnings report set to be released next week, market analyst Damir Tokic has voiced concerns about the company's future prospectsHe speculates that the earnings guidance may reveal a slowdown in performance, potentially dampening the inflated expectations surrounding the artificial intelligence sectorTokic’s multi-faceted analysis aims to shed light on the implications of these developments, particularly in the burgeoning field of generative AI.

Being bearish on Nvidia's stock rating is not a novel stance for TokicBack on June 19, 2024, when Nvidia shares hovered around $138 per share, he placed a "bearish" rating on the stockAs of now, the stock price oscillates in the same range, showing little movement over the last eight monthsThis stall in price performance is indicative of broader market questions surrounding Nvidia's valuation and the sustainability of its growth trajectory.

The past eight months have been anything but stable for NvidiaIn August, the stock plummeted to $91 before staging a robust rebound to $150, only to settle back down to approximately $133 recentlyWith the earnings report due out on February 26, Nvidia's share price has surged again, nearing its historic high of just above $150. Nevertheless, Tokic observes that the formation of a market peak may be in the works, especially given the overall long-term trends suggested by market data.

Tokic's bearish rating was primarily influenced by two considerable factors: Nvidia's stock was significantly overvalued, and its revenue growth was beginning to slowIndeed, when reviewing the revenue figures, it's clear the trajectory has falteredFrom the start of 2024, Nvidia's quarterly revenue growth percentage nosedived from a staggering 265% year-on-year to just 93.61%. Additionally, starting in April 2024, the year-on-year growth rates of net margins dwindled from 339% down to just 95%. For the approaching fourth quarter, Nvidia's anticipated revenue is expected to come in at around $38.09 billion, demonstrating a year-on-year increase of 72%, but yet it signals a continuation of a declining growth trend.

In June 2024, Nvidia's price-to-sales ratio soared to an astonishing 40 times—a dangerously high figure

Advertisements

Even though it's decreased to 30 times now, it still remains in the high territory that raises red flags for potential investorsDriven by the ongoing decline in revenue growth and this inflated price-to-sales ratio, Tokic holds firmly to the notion that a bearish perspective on Nvidia remains rational.

When we look ahead to the earnings announcement scheduled for the evening of February 26, 2024, many analysts are projecting that Nvidia could see earnings of $0.84 per share, reflecting a marginal increase of about 3.7% alongside anticipated revenues of approximately $88.09 billion, which would mark an increase of about 8.5%. However, given the stringent price-to-sales ratio that suggests higher market expectations, such growth figures would not be particularly impressive.

Historically, Nvidia has typically exceeded its revenue estimates by roughly $2 billion, consistently beating earnings expectations and upwardly revising guidance to reflect a more optimistic outlook for the following quarterThis pattern of exceptional performance has set a precedent that investors have come to expect.

Based on the strong overall performance of S&P 500 companies in the fourth quarter of 2024, Tokic predicts that Nvidia's earnings and revenue will indeed surpass expectations, albeit to a lesser degree than in previous quartersThis cooling trend, he believes, has become a constant across the boardHowever, he warns that the earnings guidance may surprise the market, particularly with the technological advancements brought forth by DeepSeek that could fundamentally alter the landscape of the AI sector.

Specifically, DeepSeek’s advancements, which utilize Nvidia's relatively lower-tier chips in smaller quantities, have led to the development of a highly competitive large language model at a fraction of the cost compared to similar models built by major American tech firmsThis presents a significant challenge to Nvidia, as the number of chips needed for generative AI models is substantially lessened, thus reducing reliance on advanced chips and potentially leading to diminished capital expenditures in the GenAI space, directly impacting Nvidia's revenue.

The breakthrough by DeepSeek occurred in late January, coinciding with the time frame where major American tech companies—such as Microsoft, Meta, and Alphabet—were preparing to share their earnings data

Advertisements

Tokic posits that due to the timing, these corporations may have been unable to adjust their earnings reports to reflect the changes in anticipated capital expenditures for generative AI that DeepSeek’s emergence has brought forward.

As major tech players come to realize the profound impact of the DeepSeek event, it can be expected that the capital expenditures in the generative AI sector will struggle to maintain previous explosive growth levels heading into 2025. As Nvidia presents its financial report, a significant focus will likely be placed on this dramatic shiftTokic argues that the company will have to address the ramifications of DeepSeek, which may lead to earnings guidance falling below market expectations and potentially result in zero growth for the quarter.

From a macro perspective, the narrative surrounding the generative AI bubble finds its origins in Nvidia’s report from April 2023, wherein revelations surfaced regarding substantial capital investments from hyperscale data center operators in the sectorTokic asserts that if these major players decide to cut back on capital expenditures, the bubble could burst at any momentThis potential decline can be traced back to the relatively low return on investment attributed to GenAI, prompting eventually reduced company investments, coupled with the impact of DeepSeek, which is likely to instigate a more immediate response from businesses.

The possibility of Nvidia revealing during its earnings call that this quarter has indeed seen a slowdown in capital expenditures for generative AI could mark a definitive breaking point for the bubbleThe discussion of these negative outlooks is already permeating the market and drawing attention even from high-profile figures like Fed Vice Chairman Philip Barr, who has reflected on the possible overhyping of generative AI technologiesBarr remarked, "As I have stated before, if productivity can see widespread incremental increases, then the economy will follow suit

Advertisements

However, currently, the value anticipated from generative AI may indeed be overestimated—a longing for revolutionary breakthroughs might ultimately culminate in only minor productivity enhancementsShould reality diverge from these expectations, companies that have heavily invested in this technology sector may find their market valuations subjected to adjustments."

In summary, combining various analyses, Tokic surmises that even without considering the potential negative impacts of the DeepSeek events, Nvidia’s growth rate is undeniably slowing, while its valuation remains in a bubble territory, marking a clear rationale for a bearish outlook.

However, when factoring in the ramifications of DeepSeek, the case for bearish sentiment on Nvidia becomes particularly compellingTokic anticipates that Nvidia will acknowledge the adverse consequences of DeepSeek in their earnings report, leading to guidance that may not meet market expectations—this is poised to play a significant role in bursting the generative AI bubble.