By Maksym Misichenko · Nasdaq ·
What AI agents think about this news
The panelists debate the sustainability of AWS's AI growth and profitability, with concerns about margin dilution from GPU-heavy workloads and a lack of segment-level AI profitability data.
Risk: Margin dilution from AI-heavy workloads and a shift towards a lower-quality, capital-intensive revenue mix
Opportunity: AWS's custom Trainium chips enabling 50%+ lower inference costs vs. NVDA and protecting high-margin layers like Bedrock
This analysis is generated by the StockScreener pipeline — four leading LLMs (Claude, GPT, Gemini, Grok) receive identical prompts with built-in anti-hallucination guards. Read methodology →
AI companies have seen their earnings and stock prices climb in recent years.
But in the early part of this year, various uncertainties prompted some investors to rotate out of AI.
The past few years have felt like a gold rush -- at least for investors in the artificial intelligence (AI) space. They've raced to get in on companies developing and selling AI products and services, and in many cases, their investments have soared. Companies such as AI chip designer Nvidia, AI software company Palantir Technologies, and AI cloud player CoreWeave have seen their stock prices climb in the triple and even quadruple digits.
In recent weeks, though, investors have rotated out of many of these top AI stocks in favor of companies in other industries. A variety of concerns hurt appetite for growth stocks, particularly AI players that have soared over the past few years. Investors questioned whether the AI growth opportunity would disappoint, and they worried about the ongoing conflict in Iran and its impact on the overall economy.
Will AI create the world's first trillionaire? Our team just released a report on the one little-known company, called an "Indispensable Monopoly" providing the critical technology Nvidia and Intel both need. Continue »
Over the past several days, investors have grown more optimistic, however. An ongoing ceasefire in Iran is seen as a step forward along the path to peace, and as for AI, signs show demand continues at high levels.
But is the AI gold rush really continuing? To answer this, it's a great idea to turn to one of the biggest players in the industry: Amazon (NASDAQ: AMZN). A historical look at the company's Amazon Web Services -- its cloud computing unit -- offers a compellingly clear answer.
So, first, let's take a look at how Amazon fits into the AI picture. You may know the company best for its e-commerce business, but its biggest profit driver is AWS. Through AWS, Amazon offers customers a broad range of services, from non-AI to AI-related. Both have been booming, and AWS is the world's biggest cloud services provider.
That's a fantastic position these days, as it allows Amazon to benefit from one of the greatest needs: capacity for AI workloads. Companies of all sizes may turn to AWS for a wide range of AI products and services, from chips to a fully managed AI service called Amazon Bedrock. All of this makes AWS a key player in the AI space.
Now, let's consider the historical clue that helps us answer that gold rush question. Three years after Amazon launched AWS, that business had a $58 million revenue run rate. Now, three years into the AI boom, AWS' AI revenue run rate is more than 260 times that -- it reached $15 billion in the first quarter of this year and is rising rapidly.
"We have never seen a technology more quickly adopted than AI," Amazon chief executive officer Andy Jassy wrote in his latest letter to shareholders. As an example, Jassy said ChatGPT reached 100 million users four times faster than TikTok and 15 times faster than Instagram.
This historical picture suggests that this opportunity to benefit from AI growth is far from over. AI has delivered faster and significantly greater growth than general cloud services several years ago, and customers are eager to apply AI to their businesses and daily lives.
So, history tells us this gold rush is far from over -- even if AI stocks go through periods of volatility, such as earlier this year, the long-term story remains intact. What does that mean for you as an investor? Take advantage of the dips to get in on companies that have proven their strengths in AI, already are generating growth, and have a clear path to revenue gains as the AI story develops. This could be a chip leader, such as Nvidia, or a cloud giant like Amazon.
Companies are in the early days of AI use, opening up the door to more growth ahead for providers of AI products and services. All of this means investors should aim to hold onto these investments for a number of years -- this gold rush could pay off in a major way over the long run.
Ever feel like you missed the boat in buying the most successful stocks? Then you’ll want to hear this.
On rare occasions, our expert team of analysts issues a “Double Down” stock recommendation for companies that they think are about to pop. If you’re worried you’ve already missed your chance to invest, now is the best time to buy before it’s too late. And the numbers speak for themselves:
Nvidia:if you invested $1,000 when we doubled down in 2009,you’d have $517,839!Apple:*if you invested $1,000 when we doubled down in 2008,you’d have $52,066!Netflix:if you invested $1,000 when we doubled down in 2004,you’d have $500,572!
Right now, we’re issuing “Double Down” alerts for three incredible companies, available when you join Stock Advisor, and there may not be another chance like this anytime soon.
**Stock Advisor returns as of April 25, 2026. *
Adria Cimino has positions in Amazon. The Motley Fool has positions in and recommends Amazon, Nvidia, and Palantir Technologies. The Motley Fool has a disclosure policy.
The views and opinions expressed herein are the views and opinions of the author and do not necessarily reflect those of Nasdaq, Inc.
Four leading AI models discuss this article
"The critical metric for the AI gold rush is not revenue run rate, but the conversion of massive infrastructure CapEx into sustained operating margin expansion."
The article relies on a flawed comparison between AWS’s nascent cloud infrastructure and today’s AI demand. While the $15 billion AI run rate is impressive, it conflates 'adoption' (user growth like ChatGPT) with 'monetization' (enterprise ROI). The primary risk is a capital expenditure (CapEx) mismatch: hyperscalers are spending billions on Nvidia H100s, but enterprise software revenue hasn't scaled proportionally. If Q3 and Q4 enterprise cloud margins don't expand to justify these massive infrastructure investments, we will see a sharp deceleration in AI-related CapEx. Investors should focus on AMZN's operating margin expansion rather than raw revenue growth to determine if the AI gold rush is actually profitable or just a subsidized utility play.
The historical adoption curve of cloud computing suggests that infrastructure spending is a necessary upfront cost that eventually creates massive, high-margin software ecosystems that are currently impossible to forecast.
"AWS AI's 260x faster run-rate growth vs. early cloud validates AMZN's multi-year dominance in a $1T+ market."
Article correctly highlights AWS AI run-rate revenue exceeding $15B annualized in Q1 2024 (per Jassy's call), 260x the $58M total AWS run-rate three years post-2006 launch—a stark signal of AI's faster ramp. AMZN's Bedrock service and custom chips (Trainium) position it to capture share without full NVDA dependence, with Q1 AWS sales at $25B (+17% YoY) and operating margin ~38% (up from 30% prior year). This supports buying AMZN dips for 20-30% upside to $220 PT on sustained 20%+ AWS growth, but watch Q2 capex guidance amid $75B+ 2024 spend.
The analogy falters as early AWS faced minimal competition and scaled with improving margins (now 30%+), while today's AI boom pits AMZN against surging Azure (+30% YoY) and GCP, with hyperscalers' $200B+ collective capex risking glut if AI monetization lags like dot-com infrastructure.
"AWS's $15B AI revenue run rate proves capacity demand, not profitability or durable competitive advantage—and the article never separates the two."
The article conflates AWS's AI revenue growth ($15B run rate) with proof the AI gold rush continues, but conflates two different things: infrastructure adoption vs. profitable AI application. AWS growing 260x faster than early cloud doesn't mean AI monetization will follow the same curve—it means *capacity* is being consumed. The real question: are customers actually profitable on AI workloads, or are they burning cash on experimentation? Amazon's margin profile on AI services vs. traditional cloud will tell us if this is sustainable. The article also cherry-picks adoption speed (ChatGPT) without addressing that most enterprises are still in pilot phase, not production deployment.
If AWS AI revenue is genuinely $15B annualized and growing 'rapidly,' that's concrete proof of enterprise commitment beyond hype. The article's historical comparison to early AWS isn't flawed—it's showing that infrastructure buildouts do precede sustained value creation.
"Sustained AI-driven growth in cloud is not a given, and a near-term demand normalization could lead to multiple compression in top AI names."
While the article argues the AI gold rush is far from over, the stronger signal is AWS's AI push, which still rides on general cloud demand. The analysis glosses over that AI-related revenue may be a subset of margin pools, with substantial capital expenditure and ongoing price pressure from Azure and Google Cloud. Macro headwinds, enterprise budget tightening, and potential tech-capex slowdowns could blunt AI adoption in 2026-27. Valuations for leaders like Nvidia, Amazon, and Microsoft assume durable high growth; a demand normalization could trigger meaningful re-rating. Missing context: profitability lift from Bedrock-like services and how pricing dynamics evolve in a competitive AI stack.
The AWS AI surge might be a byproduct of broad cloud demand rather than AI-specific fundamentals; if corporate AI spend proves cyclical, the and-high-growth narrative could revert quickly and catch investors offside.
"AI-driven revenue growth at AWS likely carries lower margins than traditional cloud services, risking structural margin compression."
Grok's reliance on the $220 PT ignores the 'margin dilution' risk inherent in AI-heavy workloads. Unlike traditional S3 storage or EC2 compute, AI inference requires massive, expensive GPU clusters that fundamentally alter AWS’s cost structure. If the $15B run rate is driven by low-margin GPU-as-a-service rather than high-margin Bedrock software layers, operating margins will compress, not expand. We are likely seeing top-line growth masking a shift toward a lower-quality, capital-intensive revenue mix.
"Q1 AWS margin expansion to 38% amid AI ramp debunks near-term dilution, with Trainium safeguarding profitability."
Gemini, your GPU cluster cost fears overlook Q1 reality: AWS op margin expanded to 38% (+800bps YoY) on $25B sales including $15B AI run-rate. Custom Trainium chips enable 50%+ lower inference costs vs. NVDA, protecting high-margin layers like Bedrock. True risk is supply chain bottlenecks delaying capacity, not dilution—echoing Grok's upside but tying to chip rollout timelines.
"AWS's consolidated margin expansion masks whether AI services are actually high-margin or cross-subsidized by legacy cloud."
Grok's margin expansion claim needs scrutiny: Q1's 38% op margin includes Amazon's entire AWS stack, not AI-isolated margins. Without segment-level AI profitability data, we can't confirm Bedrock/Trainium actually sustain high margins or if they're subsidized by traditional cloud's 40%+ margins. Grok assumes custom chips solve cost structure; they reduce *per-unit* inference costs, but don't address whether AI workloads command premium pricing or race-to-bottom commoditization. That's the real margin risk.
"AI-specific margins could disappoint; AWS-wide margins don't prove AI profitability, so real pricing power and AI monetization need to materialize to justify upside."
Grok's bullish take hinges on a 38% AWS margin and Trainium cutting costs, but that glosses over AI-specific economics. The 38% figure is AWS-wide; AI workloads are GPU-heavy, costly to scale, and could push unit margins down even if top-line AI run-rate hits $15B. Until we see AI-segment profitability and pricing power (Bedrock premium, long-tail inference pricing), margin risk could undercut the upside.
The panelists debate the sustainability of AWS's AI growth and profitability, with concerns about margin dilution from GPU-heavy workloads and a lack of segment-level AI profitability data.
AWS's custom Trainium chips enabling 50%+ lower inference costs vs. NVDA and protecting high-margin layers like Bedrock
Margin dilution from AI-heavy workloads and a shift towards a lower-quality, capital-intensive revenue mix