Is NVIDIA Corporation (NVDA) a Good Stock to Buy Now? Bull and Bear Cases Examined
Key Takeaways
- •NVIDIA reported $82 billion in revenue for the latest period, representing 85% year-over-year growth and marking its fourteenth consecutive quarter of sequential revenue increases.
- •Investor Michael Burry has outlined a bearish thesis comparing hyperscaler AI infrastructure spending to the late-1990s Dot-Com bubble, warning that lagging enterprise monetization could trigger severe capital expenditure cuts.
- •NVIDIA's current stock valuation implies the company would need to generate approximately $250 billion in net profits annually for the next two decades to justify its share price.
- •The number of hedge fund portfolios holding NVDA rose to 275 at the end of the first quarter, up from 264 in the prior quarter, indicating growing short-term bullish institutional positioning.
- •NVIDIA issued second-quarter revenue guidance of $91 billion while maintaining gross margins of approximately 75% despite a rapid production ramp of its Blackwell architecture.

Is NVIDIA Corporation (NVDA) a Good Stock to Buy Now? Bull and Bear Cases Examined
NVIDIA Corporation (NVDA) shares were trading at $206.84 as of July 24, with trailing and forward price-to-earnings ratios of 32 and 23.5, respectively, according to Yahoo Finance. A bullish thesis on NVIDIA published on Compounding Your Wealth's Substack by Sergey argues the company has evolved well beyond a traditional semiconductor manufacturer into a full-stack AI infrastructure platform encompassing GPUs, CPUs, networking, systems, CUDA software, AI libraries, and a deeply integrated partner ecosystem. The debate over NVIDIA's trajectory comes at a time when global AI infrastructure spending has reached unprecedented levels, with hyperscalers including Microsoft, Google, Meta, and Amazon collectively committing tens of billions in annual capital expenditure toward AI compute capacity.
The Bullish Case: Vertical Integration and Platform Dominance
NVIDIA remains the dominant supplier of AI compute infrastructure on a global scale. As AI data centers increasingly prioritize token throughput, cost per token, utilization, uptime, and software efficiency over raw standalone hardware performance, the company's vertically integrated platform has strengthened its competitive position. While AMD has positioned its MI300X accelerators as the primary competing AI GPU and hyperscalers such as Google (TPU), Amazon (Trainium), and Microsoft (Maia) have advanced internal custom-silicon programs, NVIDIA's CUDA software ecosystem — cultivated over more than a decade — remains a deep moat that no competitor has yet replicated at comparable scale.
NVIDIA's Blackwell architecture has been broadly deployed across hyperscalers, frontier AI model developers, AI cloud providers, sovereign AI initiatives, and enterprise AI infrastructure. Management noted adoption by every major hyperscaler and leading model builder. Networking has also emerged as a meaningful growth engine — Spectrum-X has surpassed the combined scale of competing Ethernet networking peers, while InfiniBand revenue increased more than fourfold year over year.
The company reported exceptional financial results. Revenue reached $82 billion, representing 85% year-over-year growth and 20% sequential growth — marking NVIDIA's fourteenth consecutive quarter of sequential increases. Data center revenue climbed 92% to $75 billion, driven by $60 billion in compute revenue and $15 billion in networking revenue. Hyperscalers accounted for roughly half of data center sales.
Despite the rapid Blackwell production ramp, gross margins held at approximately 75%. Free cash flow reached a record $49 billion, and the company returned $20 billion to shareholders while authorizing an additional $80 billion share repurchase program.
NVIDIA's system-level integration across Grace Blackwell, Vera CPUs, NVLink, Spectrum-X, InfiniBand, CUDA, and AI software creates a competitive moat that is difficult for rivals to replicate. MLPerf benchmarks demonstrated Blackwell Ultra's advantages, delivering 2.7 times higher throughput while reducing cost per token by 60%.
According to the bulls, although supply constraints, China-related risks, customer concentration, and custom AI silicon remain key challenges, NVIDIA has proactively expanded supply commitments, inventory, and prepayments to support future demand. U.S. export controls on advanced semiconductors have restricted NVIDIA's ability to sell its highest-performance products in China, a market the company has historically addressed with modified chip variants, adding a layer of geopolitical uncertainty to its growth outlook. With second-quarter revenue guidance of $91 billion, stable margins, the upcoming Vera Rubin platform, and continued AI infrastructure investment, bulls argue the company is well positioned for sustained growth — provided enterprise AI spending continues translating into profitable customer deployments.
The Bearish Case: Michael Burry's Cyclical and Macro Concerns
Not all investors are optimistic. Investor Michael Burry has outlined a bearish thesis built on a macro market-bubble framework combined with classic semiconductor cyclicality concerns. The semiconductor industry has a well-documented history of boom-bust cycles, with severe downturns in 2001, 2008, and 2018–2019 driven by inventory corrections and demand collapses, providing historical precedent for the cyclical risks Burry highlights.
At the macro level, Burry views the aggressive multi-billion-dollar infrastructure spending by mega-cap technology hyperscalers as a classic speculative mania, drawing direct parallels to the late-1990s Dot-Com bubble. His core argument holds that enterprise monetization and end-user return on investment from software applications will inevitably lag far behind the front-loaded costs of hardware deployment. When hyperscalers realize their current infrastructure yields diminishing marginal returns, capital expenditure budgets would face severe rationalization, exposing the semiconductor supply chain to an abrupt demand decline.
From a microeconomic perspective, Burry treats NVIDIA as a hardware provider subject to the classic "bullwhip effect" that has historically plagued semiconductor cycles. During periods of extreme chip supply constraints, enterprise customers typically double-order GPUs to secure allocation. Once supply catches up with actual immediate utilization, buyers pause purchases to absorb excess inventory, leading to sharp order cancellations and inventory drawdowns. Burry further argues that NVIDIA's record-high gross margins are cyclically unsustainable — as custom enterprise silicon and competing hardware options mature, pricing power would normalize toward historical industry averages.
Valuation Context and Hedge Fund Positioning
A prior bullish thesis on NVIDIA by Compounding Your Wealth was covered in April 2025, highlighting the company's AI hardware leadership, CUDA ecosystem, expanding data center opportunity, and attractive valuation despite moderating growth. Since that coverage, NVDA's stock price has appreciated approximately 90.17%.
The current valuation, however, implies that NVIDIA will need to generate around $250 billion in net profits every year for the next two decades to justify its share price — a notably high bar over an extended time horizon, and one that presupposes the AI infrastructure buildout continues at or near current rates without a material cyclical interruption.
NVIDIA ranks among the 40 Most Popular Stocks Among Hedge Funds. According to the cited database, 275 hedge fund portfolios held NVDA at the end of the first quarter, up from 264 in the previous quarter, indicating short-term bullish positioning among institutional investors.
Disclosure: None.