After years of relative quiet, the e-commerce and cloud computing giant Amazon.com (AMZN.US) is making a powerful return to the forefront of the global technology stage. Investor optimism regarding the company's increasingly solid position in the artificial intelligence (AI) field is fueling a sustained surge in its stock price, propelling its market capitalization past key milestones. From strategic investments to in-house hardware development, from deep partnerships with large language model (LLM) companies to comprehensively empowering its core businesses, Amazon is undergoing a comprehensive value reassessment centered on "AI+."

Investor enthusiasm for the e-commerce and cloud giant's AI strategy continues to build, driving its stock price up significantly since late March and pushing its market cap toward the historic $3 trillion threshold. Latest data shows Amazon's stock trading around $270, with a market capitalization of approximately $2.91 trillion. It now needs only about a 2% gain to join the exclusive $3 trillion club, whose current members are only Microsoft, Apple, NVIDIA, and Alphabet.

Since hitting a low on March 27, Amazon's stock has surged 36%, making it the fourth-largest contributor to gains in the S&P 500 index, accounting for 7.4% of the index's 17% rise over the same period. Its 27% surge in April alone marked its best monthly performance since 2007. Year-to-date in 2026, the stock is up approximately 18.7%, significantly outperforming the Roundhill Magnificent Seven ETF, which tracks the "Magnificent Seven" tech stocks, as well as the broader market.

It is worth noting that Amazon endured a five-year period of low returns prior to this rally. If the current uptrend continues and it successfully surpasses $3 trillion, Amazon will have tripled its market cap in roughly six years and three months—it first reached $1 trillion in February 2020, broke $2 trillion in June 2024, and the journey to the third trillion is progressing far faster than the first two.

**Strong Fundamental Support: AWS Returns to High Growth, Chip Business Emerges**

This sharp rally is not without foundation; it is backed by robust performance from the company's core businesses, particularly in cloud computing and AI-related segments. The first-quarter 2026 earnings report released on April 29 served as a key catalyst.

**AWS Growth Hits 15-Quarter High** The report showed Amazon's Q1 net sales reached $181.5 billion, a 17% year-over-year increase, exceeding market expectations. The most closely watched segment, cloud business AWS, achieved revenue of $37.6 billion, a sharp 28% increase year-over-year, marking its fastest growth rate in 15 quarters. AWS operating profit reached $14.2 billion, accounting for nearly 60% of the company's total operating profit, highlighting its powerful profitability.

On the earnings call, CEO Andy Jassy used a striking comparison to illustrate the explosive speed of the AI business: "Three years after AWS launched, its annualized revenue was $58 million. In the first three years of this AI wave, AWS's annualized AI revenue has already exceeded $15 billion, a scale nearly 260 times larger."

AWS's backlog of commitments also provides a vision for sustainable growth. As of the end of Q1, AWS's backlog stood at $364 billion, a figure that does not yet include a new agreement worth over $100 billion recently reached with Anthropic, representing a significant increase from $244 billion at the end of 2025. Jassy emphasized that the backlog is not concentrated with one or two large customers but has "a fairly broad customer composition."

**Dual Engines: Custom Chips and Advertising** More exciting for the market is the first systematic disclosure of the scale of Amazon's in-house AI chip business. CEO Andy Jassy revealed that the company's chip business (encompassing Graviton, Trainium, etc.) has achieved an annualized revenue exceeding $20 billion and is poised for triple-digit percentage year-over-year growth. Meanwhile, the advertising business surpassed $70 billion in revenue over the past twelve months, becoming another powerful growth engine for Amazon.

Amazon deployed over 2.1 million AI chips in the past twelve months, with more than half being its in-house Trainium chips. Jassy presented a figure that caught the market's attention: "If our chip business existed as a standalone company, selling the chips it produces to AWS and third parties like other leading chip companies do, our annualized revenue would be $50 billion."

Progress on the Trainium series is rapid: Trainium 2 offers approximately a 30% better price-performance ratio compared to similar GPUs and is now essentially sold out; Trainium 3 began shipping in early 2026. Amazon has secured over $225 billion in revenue commitments for Trainium chips, primarily from deep partnerships with Anthropic and OpenAI, with most of the chip production capacity already booked.

In the e-commerce domain, AI is also driving a profound transformation. On May 13, Amazon announced it would phase out the standalone Rufus shopping chatbot, instead deeply integrating AI shopping functionality into the Alexa assistant, launching a new "Alexa for Shopping." This integration means Amazon's search box will transform into a question-and-answer engine, allowing users to obtain product information, compare items, and even schedule automatic purchases when specific prices are met through natural language conversation.

Daniel Rausch, head of Amazon's Alexa business, commented on this: "Shopping is not something you can treat as a side hustle." He emphasized that Amazon's unique data—customer reviews, vast product catalog, inventory status, and estimated delivery dates—constitutes a competitive advantage that other AI shopping tools cannot replicate.

The strategic significance of this move is that AI is not only enhancing AWS's growth momentum but is also reshaping the user experience and commercial monetization logic of Amazon's core e-commerce business. Search behavior is shifting from keyword matching to intent understanding, ad placement is moving from bid-based ranking to contextual embedding, and the traffic acquisition methods for millions of third-party sellers will undergo fundamental changes.

**Dual Bets: Deep Partnerships with Anthropic and OpenAI** In the field of large AI models, Amazon has adopted a "dual bet" strategy, and the stakes on both sides are continuously increasing. On April 20, Amazon announced an additional $5 billion investment in OpenAI's rival Anthropic, with the potential to add up to $20 billion more based on commercial milestones. Combined with the cumulative $8 billion invested since 2023, the total committed investment is substantial.

In exchange, Anthropic committed to investing over $100 billion in AWS technology over the next decade, covering current and future generations of Trainium chips and tens of millions of Graviton CPU cores. Anthropic will gain access to Trainium chips with a total computing power of up to 5 gigawatts for training and powering its AI models.

While securing a future supply of AI models through Anthropic, Amazon also announced a multi-year strategic partnership with OpenAI in February, investing $50 billion in the latter. With this, the world's two most influential AI model companies are now part of the AWS ecosystem.

Amazon Web Services announced in late April the deepening of its partnership with OpenAI, introducing the latest flagship models like GPT-5.5 and GPT-5.4 to the Amazon Bedrock platform and launching products like the OpenAI-based programming assistant Codex. This series of strategic moves breaks the long-standing exclusive cloud service arrangement between Microsoft Azure and OpenAI, importing a richer model ecosystem and developer resources into AWS.

Stephen Lee, founding partner of Logan Capital Management, commented on this: "All of Amazon's businesses are interconnected. Mastering AI is not only beneficial for AWS but also brings huge advantages to logistics and ad targeting for its e-commerce business. Amazon should become a dual winner in both AI infrastructure and AI applications—a highly attractive combination."

**The Computing Arms Race: A $200 Billion Gamble and a "Once-in-a-Lifetime Opportunity"** Amazon's ambition extends far beyond current partnerships. In the arms race for AI infrastructure, it is making an exceptionally large bet in tech history. Amazon expects its total capital expenditures for 2026 to reach approximately $200 billion, a surge of over 50% year-over-year. This figure far exceeds previous Wall Street expectations and makes it the S&P 500 component company with the highest capital expenditures.

CEO Andy Jassy was firm in the face of Wall Street skepticism, stating bluntly in his annual shareholder letter: "We are not going to take a conservative approach on this—we are going to invest to become a truly meaningful industry leader." He even defined AI as a "once-in-a-lifetime opportunity."

The vast majority of this massive spending will be directed toward AI infrastructure, and the commercial prospects of its custom AI chip, Trainium, are central to supporting this investment logic. According to Jassy's latest disclosure, Amazon plans to not only use Trainium for its own cloud services within two years but also to sell it directly to external customers, directly challenging NVIDIA's dominance in this field.

The latest generation Trainium 3 chip is now commercially available, offering double the performance of its predecessor, while the more advanced Trainium 4 is already almost entirely booked by customers.

**P/E Ratio Under 25x, a Valuation Bargain? Wall Street's Bullish Consensus and Concerns** As earnings expectations are revised upward, Amazon's valuation appears increasingly attractive. Over the past month, the consensus estimate for Amazon's 2026 earnings per share has been raised by 14%, and revenue expectations have also increased. This puts Amazon's current price-to-earnings (P/E) ratio at under 25 times, significantly below its 10-year average of 46 times. In late March, its P/E ratio had briefly fallen to its lowest point since the end of 2008.

This characteristic of the stock becoming "cheaper as it rises" largely explains why Wall Street assigns Amazon the highest proportion of "Buy" ratings among large-cap stocks. Of the 83 analysts tracked by institutions, 79 have a "Buy" rating, with none issuing a "Sell" rating. The average price target is $313, implying roughly 16% upside potential over the next 12 months.

Facing such an aggressive investment strategy, the prevailing sentiment on Wall Street is currently one of approval. Among tracked analysts, bullish sentiment is extremely high, with an average price target around $313, suggesting about 15% upside over the next year. Multiple investment banks, including JPMorgan, Piper Sandler, Canaccord, and UBS, have raised their price targets for Amazon following the Q1 earnings report. TD Cowen reiterated its Buy rating and maintained a $350 price target.

Over the past month, the consensus estimate for Amazon's 2026 EPS has been raised by 14%, and its revenue expectations have also increased. This makes Amazon's current P/E ratio of under 25 times appear quite attractive—a figure well below its average over the past decade and one that had previously hit a low in late March not seen since the end of 2008.

However, behind the surging market sentiment, a core question remains unresolved—what returns can Amazon's massive 2026 capital expenditure plan of approximately $200 billion actually deliver? In the first quarter, Amazon's capital expenditures reached $44.2 billion, primarily for AWS and generative AI computing capacity build-out.

A direct consequence of this massive investment is a sharp deterioration in free cash flow—over the past twelve months, free cash flow was only $1.2 billion, a 95% plunge from $25.9 billion a year earlier. Amazon expects full-year 2026 capital expenditures of about $200 billion, significantly higher than the $131 billion in 2025, with 2027 expenditures projected to climb further to $226 billion.

The combined capital expenditures of Amazon, Microsoft, Alphabet, and Meta for 2026 exceed $700 billion. According to Goldman Sachs calculations, by then, almost all the free cash flow of the "Magnificent Seven" tech giants will be consumed by capital expenditures.

Jassy attempted to soothe market concerns in his April 9 shareholder letter: For the 2026 capital expenditures, most of the capacity will begin generating revenue in 2027-2028, and "a significant portion is already backed by customer commitments." In other words, this is not a blind gamble; buyers are already in line.

Not everyone is confident in this massive bet. Tom Graff, Chief Investment Officer at Facet, expressed a cautious view. Although his firm still holds Amazon stock, he believes "there is still a huge question about how much return all this AI spending will ultimately bring." He noted that as long as capital expenditures continue to soar, the company's valuation multiple could face pressure because it will no longer have the strong free cash flow it once did. "Ultimately, I think it underperforms more than it outperforms," Graff added. "A lot of things have to go exactly right, and in a world where I'm trying to balance risk and reward, I see more risk."