July 6, 2026, marked a strong rebound for the US semiconductor sector. The Philadelphia Semiconductor Index surged more than 4% in a single day, ending a two-day losing streak. As of the close on July 7 (Beijing time), the Nasdaq Composite rose 288.49 points, or 1.12%, to finish at 26,121.16. The S&P 500 climbed 0.72% to 7,537.43, while the Dow Jones Industrial Average gained 0.29% to close at 53,055.91—breaking the 53,000 mark for the first time and setting a new all-time high.
During this rally, Advanced Micro Devices (AMD) jumped 6.61% to $552.05, bringing its year-to-date gains to 157.77%. The main catalyst behind this surge was a Goldman Sachs research report released on the evening of July 5, which raised AMD’s 12-month price target from $450 to $640 and maintained a "Buy" rating.
This target hike represents a 42% increase, making it one of the most aggressive calls among major Wall Street investment banks in the current cycle. Meanwhile, NVIDIA closed at $195.55 on July 7, up 0.37%. The valuation gap and product narratives between these two AI chip giants are sparking widespread debate in the market.
What does Goldman Sachs’s $640 price target really mean? How far can AMD’s AI chip story go? This article will offer a structured analysis from four perspectives: the rationale behind the price target increase, an objective review of Goldman Sachs’s historical accuracy, the competitive landscape between AMD and NVIDIA, and the beneficiaries across the industry supply chain.
Goldman Sachs Significantly Raises AMD Price Target: The Underlying Logic
Goldman Sachs analyst James Schneider outlined three main reasons for the price target upgrade in his report.
First, persistent demand for CPUs driven by Agentic AI. As AI applications shift from training to inference, the role of CPUs in AI infrastructure is being reevaluated. Goldman Sachs believes AMD’s expanding share in the server CPU market will deliver revenue that exceeds expectations.
Second, continued growth in overall AI infrastructure spending. Global cloud service providers remain in an expansionary capital expenditure cycle, providing structural support for AMD’s data center business.
Third, strong Q2 earnings expectations. Goldman Sachs predicts AMD will deliver a "strong" second-quarter report and robust guidance in August. Schneider specifically noted that AMD’s comments on new customer engagement could boost the 2027 valuation for its data center segment, with the market closely watching AMD’s collaborations with Meta and OpenAI.
It’s worth noting that this is Goldman Sachs’s second major price target hike for AMD in about two months. Previously, AMD’s share price had pulled back from its 52-week high of $584.73 to $517.82, and Goldman’s latest stance reframes this correction as a "buying opportunity" rather than a signal of a trend reversal.
Market reaction has validated this view in the short term. AMD surged as much as 10.5% intraday on July 6, ultimately closing up 6.61%.
Goldman Sachs Price Target Accuracy: A Variable to Approach with Caution
When evaluating the credibility of the $640 price target, it’s important to review Goldman Sachs’s track record as a sell-side institution. Objectively, the bank’s forecasting accuracy has not always been consistent.
On macro forecasts, Goldman Sachs has been correct in only 3 out of 8 predictions for China’s stock market, with a hit rate of about 37%. For example, in 2018, it predicted A-shares would rise 14%–17%, but the market actually fell by more than 20%.
In commodities, Goldman Sachs recommended "selling gold" in 2008, but gold prices rose 5.66% that year and soared 24.57% the next. In December 2012, it forecast gold would rise to $1,800, but six months later prices had dropped to $1,180. At the time, the market jokingly referred to Goldman as a "contrarian indicator" for gold prices.
For individual stock ratings, Goldman Sachs consistently rated Bank of Communications A-shares as "Sell" with a price target below 4 yuan, but since 2022, the stock has repeatedly broken above 6 yuan, closing at 6.88 yuan in March 2026.
Still, this doesn’t mean Goldman’s research reports lack reference value. Sell-side price target adjustments often reflect their stage-based judgments on industry trends and company fundamentals, rather than precise stock price forecasts. The $640 target, compared to the current price of $552.05, implies about 16% upside—this expected return already factors in market pricing for uncertainty.
From another angle, Cantor Fitzgerald offered an even more aggressive target, raising theirs from $500 to $700, making it Wall Street’s most bullish forecast. Goldman’s $640 sits in the upper-middle range among institutional predictions, not an isolated aggressive call.
AMD MI Series vs. NVIDIA Blackwell: The Latest Developments in the AI Chip Battle
The foundation for Goldman Sachs’s price target hike ultimately rests on AMD’s product competitiveness in the AI chip market. Right now, the competitive landscape is evolving from "one dominant player" to "two major contenders," though the gap remains substantial.
Market share: NVIDIA still dominates. According to TrendForce, NVIDIA is expected to hold about 64% of the global AI chip market in 2026, with AMD at roughly 8.6%. UBS analysts note that NVIDIA currently controls about 80% of the hardware market for data center AI performance upgrades, while AMD’s share is only 5%–7%. In the AI training chip market, SemiAnalysis reports NVIDIA’s Q1 2026 market share at 92%, and 78% for inference chips—almost unchanged from a year ago.
Product generations: NVIDIA leads, but the gap is narrowing. NVIDIA’s Blackwell platform (B200/GB200 systems) dominates 2026 shipments, while Rubin Ultra is expected to roll out gradually starting in Q4 2026. In AA-AgentPerf DeepSeek V4 Pro benchmarks, the Blackwell architecture outperforms AMD’s MI355X in per-watt efficiency for agent workloads.
But AMD is closing the gap. The Instinct MI355X, after system-level optimization, now matches the Blackwell B200 in several key metrics—MI355X features 288GB of high-bandwidth memory, more than 60% higher than B200’s 180GB.
Next-generation products: AMD Helios platform emerges as a key variable. AMD is accelerating deployment of its next-gen Helios cabinet-level AI platform, equipped with Instinct MI450X GPUs, sixth-generation EPYC CPUs, and advanced networking solutions, with mass deployment expected in the second half of 2026. The MI455X series, launched at CES 2026, offers a tenfold performance boost over MI355X, built on 2/3nm process technology, with 320 billion transistors and flagship models featuring 432GB of HBM4 memory.
UBS expects AMD’s Helios platform motherboards to ship in Q4 2026. AMD has also announced over $10 billion in investment in Taiwan’s AI ecosystem, expanding strategic cooperation with its supply chain.
Core conclusion on the competitive landscape: NVIDIA still holds significant advantages in ecosystem, software stack (CUDA), and customer loyalty, but AMD is gaining ground in inference and some training scenarios through cost-effectiveness, open standards, and memory architecture innovation. Goldman’s price target hike is based on optimism about this trend—AMD’s potential for market share gains during the AI infrastructure spending expansion cycle is underestimated.
AMD Supply Chain Beneficiaries
The ramp-up in AMD’s AI chip production directly benefits the entire supply chain, from wafer foundries to server assembly. Here’s a breakdown of the key beneficiaries:
Wafer foundry: TSMC is the primary beneficiary. AMD’s high-end CPUs, GPUs, and AI chips rely heavily on TSMC’s advanced process and packaging technologies. As AMD and ARM expand their market share, more high-end CPU orders will flow to TSMC and other advanced process suppliers. The next-generation AMD EPYC processor, codenamed "Venice," is already in mass production using TSMC’s 2nm process—the first 2nm product to enter mass production in high-performance computing. On July 7, TSMC’s ADR rose 4.06% to $451.79.
Advanced packaging: ASE Technology Holding is a major beneficiary. The proliferation and complexity of chiplet architectures are driving sustained gains for advanced packaging firms like ASE. Every MI450X GPU sold by AMD leverages TSMC’s N2/N3 process capacity and ASE’s advanced packaging yields.
Server ODMs: Wiwynn, Quanta, Inventec, and others. System-level deployment of the AMD Helios platform requires collaboration with server manufacturers. Wiwynn, Wistron, and Inventec are actively helping build systems based on AMD Helios. The mass production of three major AI platforms—NVIDIA Vera Rubin, AWS Trainium 3, and AMD Helios—will drive a new wave of shipments across the supply chain, including Wiwynn, Quanta, and Foxconn.
Other key segments: These include PCB suppliers like Unimicron, cooling solution providers like Auras, server management chip suppliers like Aspeed, and test equipment vendors like Chroma and Honjin.
Bank of America has also raised valuation expectations for supply chain leaders like TSMC and ASE, noting that advanced process and packaging remain the most defensible segments in the supply chain.
Conclusion
Goldman Sachs’s upgrade of AMD’s price target to $640 reflects a broader revaluation of AI chip sector valuations. Achieving this target depends on multiple factors converging: the pace of Helios platform mass production and customer adoption, sustained CPU demand from agentic AI, and AMD’s real gains in AI training chip market share.
Current data shows NVIDIA’s dominance in the AI chip market is unlikely to be challenged in the near term—a massive gap remains between its 64% global market share and AMD’s 8.6%. However, AMD is leveraging faster product iterations, open ecosystem strategies, and deep supply chain partnerships to capture more share in the incremental market.
For investors, the significance of the $640 price target lies not in forecasting an exact endpoint for AMD’s share price, but in understanding Goldman Sachs’s framework for judging AMD’s market share potential during the AI infrastructure capital expenditure expansion cycle. As Goldman’s analyst pointed out, the August earnings call is a key near-term catalyst, while the "Advancing AI" annual conference on July 22–23 offers an early look at technological progress.
The AI chip race is far from over. Whether AMD can shift from "catching up" to "running alongside" will be tested over the next 12–18 months of product cycles.
FAQ
1. What is the basis for Goldman Sachs raising AMD’s price target to $640?
Goldman Sachs analyst James Schneider cited persistent CPU demand from agentic AI, expanding overall AI infrastructure spending, and strong second-quarter earnings expectations. Goldman predicts AMD will deliver a robust report in August and believes the market will closely monitor AMD’s collaborations with Meta and OpenAI.
2. What is the competitive landscape between AMD and NVIDIA in the AI chip market?
NVIDIA maintains absolute dominance, with a projected 64% global AI chip market share in 2026, compared to AMD’s 8.6%. In the AI training chip market, NVIDIA’s share reaches 92%. However, AMD is narrowing the gap through product iterations like MI355X and MI450X and the Helios platform, gaining competitiveness in inference and cost-effectiveness.
3. How accurate are Goldman Sachs’s price target forecasts historically?
Goldman Sachs’s forecasting accuracy is inconsistent. Statistics show only 3 out of 8 predictions for China’s stock market were correct, with a hit rate of about 37%. There have also been several directional misses in commodities and individual stock ratings. Sell-side price targets more often reflect stage-based industry judgments than precise stock price forecasts.
4. Which supply chain companies benefit from AMD’s ramp-up in AI chip production?
Key beneficiaries include wafer foundry TSMC (advanced process), packaging firm ASE Technology Holding (advanced packaging), server ODMs like Wiwynn, Quanta, and Inventec, as well as PCB suppliers like Unimicron, cooling solution providers like Auras, and server management chip suppliers like Aspeed.
5. How much upside remains for AMD’s share price?
Goldman Sachs’s $640 price target implies about 16% upside from the July 7 closing price of $552.05. Cantor Fitzgerald’s target is even more bullish at $700. Actual performance will depend on the pace of Helios platform mass production, the AI infrastructure capital expenditure cycle, and AMD’s progress in increasing its market share.
