In the fervent wave of AI enthusiasm, NVIDIA's stock price appears to follow a seasonal pattern of "autumn harvest and winter storage."

Goldman Sachs presented a tactically cautious view in a report released on August 24: despite being "very bullish" on NVIDIA's long-term growth prospects, based on historical trading pattern analysis, the chip giant's stock may struggle to outperform the broader market in the coming months.

The firm noted that NVIDIA's stock performance exhibits a regular pattern: rising in the first half of the year due to clear positive data catalysts, but facing pressure in the second half due to lack of new hard catalysts. For example, its stock surged 149% in the first half of 2024 but only gained 12% in the second half.

Analysts believe that without new quantitative data points sufficient to drive earnings expectation upgrades, market uncertainty about long-term direction (specifically 2027) will suppress stock trends through the end of 2025, similar to trading patterns in the second halves of 2023 and 2024.

For investors, this suggests NVIDIA may be entering a tactical consolidation period of "AI autumn." Goldman Sachs noted that from now until hyperscale computing customers provide formal capital expenditure guidance in January 2026, the market will closely watch three potential driving factors: any quantitative comments from hyperscale customers in October, company clarifications on the launch timing of the next-generation "Rubin" platform, and further clarity regarding China business and U.S. export controls.

However, Goldman emphasized this is merely a short-term tactical judgment. The report reiterated that based on positive capital expenditure comments from hyperscale customers and demand prospects from non-traditional clients, they remain "very optimistic" about NVIDIA's prospects for exceptional growth in 2026.

Historical Repetition? "First Half Surge, Second Half Consolidation"

Goldman's analysis reveals a distinct pattern in NVIDIA's stock performance. NVIDIA stock typically performs excellently in the first half of each year, primarily driven by hyperscale computing customers like Amazon, Microsoft, Meta, and Google releasing annual capital expenditure "quantitative guidance." These specific numbers provide the market with solid growth expectations, driving upward earnings forecasts and stock prices.

However, entering the second half, the situation often changes. Despite these major customers continuing to make "qualitative comments" about capital expenditure directions, the lack of new hard data to further enhance market expectations means other factors begin having greater impact on stock prices.

This pattern has been particularly evident over the past two years:

2024 Review: NVIDIA stock soared 149% in the first half of 2024, benefiting from better-than-expected capital expenditure guidance at the beginning of the year. However, entering the second half, market concerns about 2025 capital expenditure "peaking," competition from ASICs, and doubts about Blackwell product launch timing emerged, causing its stock to gain only 12% in the second half.

2023 Review: Similarly, after ChatGPT sparked the AI narrative, NVIDIA stock exploded 189% in the first half of 2023. But in the second half, despite fundamentals remaining strong, investors began questioning the sustainability of this spending boom, leading to valuation pressure and a narrowed second-half gain of 17%.

Three Key Variables Determine Year-End Performance

Looking ahead to the remainder of 2025, Goldman believes that without new annual guidance from major customers, NVIDIA's stock performance will depend on progress in three key variables.

First are comments from hyperscale computing customers when releasing third-quarter earnings in October. The market will seek any quantitative clues that might hint at 2026 capital expenditure scale, rather than merely directional descriptions.

Second is the launch timing of NVIDIA's next-generation "Rubin" product architecture. Any clear information about this platform's release schedule will affect market judgment of its future technological leadership and growth cycles.

Finally is clarity regarding its China business. As U.S. export control policies continue evolving, any further clarification about China market business prospects will become an important factor affecting investor sentiment. Goldman believes that without substantial progress in these areas, stock prices may face pressure due to lack of catalysts.

Despite issuing short-term warnings, Goldman has not changed its long-term positive view of NVIDIA. The report reiterated that based on positive cloud provider capital expenditure comments and demand prospects from non-traditional customers, they are "very optimistic" about NVIDIA achieving exceptional growth in 2026.

Goldman currently maintains a "Buy" rating on NVIDIA, with a 12-month price target of $200, based on normalized earnings per share of $5.75 and a price-to-earnings ratio of 35x.

AI Peer Differentiation: Broadcom Aligned, AMD and Marvell May Consolidate

Goldman's report also provided tactical outlooks for several other AI-related semiconductor companies.

For Broadcom, Goldman expects its second-half 2025 trading dynamics to be similar to NVIDIA's, as the market has largely digested its AI business guidance for fiscal 2026 and 2027. Incremental data from new XPU customers and AI networking business will be key to whether the stock can outperform the market.

For AMD, Goldman believes its data center GPU potential growth in 2026 and short-term strength in PC and server CPUs are already reflected in current stock prices. The company's planned investor day event in November will serve as a "litmus test" for examining its data center GPU revenue expectations and 2026 outlook.

For Marvell, Goldman expects its stock to maintain range-bound trading for the remainder of this year. Growth visibility from Amazon's custom computing business and Microsoft business in the second half of 2026 will be key to driving stock breakouts.