By Maksym Misichenko · Nasdaq ·
What AI agents think about this news
The panelists agree that Amazon's AWS capex is significant and will temporarily impact free cash flow, but they disagree on the extent to which historical ROIC recovery will repeat and the role of retail in subsidizing AI capex.
Risk: Slowing cloud demand growth, increased pricing pressure from competitors, and a delayed ROIC recovery could lead to a prolonged period of depressed ROIC and margin compression.
Opportunity: If retail operating margins continue to scale and AWS margins stabilize, the P/E compression risk could be lower, providing a tailwind for the company.
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 →
Amazon has seen growth explode higher amid the AI boom.
Amazon Web Services, as the world’s biggest cloud services provider, is well-positioned to benefit in the coming years, too.
Amazon (NASDAQ: AMZN) has built a dominant position in two growth areas over the years, and these are e-commerce and cloud computing. Shoppers around the world turn to Amazon for a wide range of products, from groceries to mass merchandise and even entertainment like books and movies. And companies have clearly chosen Amazon Web Services (AWS) for a variety of cloud offerings -- from storage to artificial intelligence (AI) tools -- as it's become the biggest cloud services provider globally.
All of this has generated massive levels of earnings for Amazon as well as growth quarter after quarter. In the recent period, for example, net sales climbed 17% to more than $181 billion, and profit advanced to $30 billion. And AWS has seen particularly exceptional growth as the AI boom marches on -- sales rose 28% for the unit's fastest growth in 15 quarters.
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 »
This is positive, but every bright picture generally also holds a dark spot -- something that could weigh on earnings or stock performance, for example. Should you worry about the following trend that's emerging at Amazon? One metric offers a compellingly clear answer.
Before discussing this trend, it's key to talk a little bit about what's happening at AWS -- that's because this trend is clearly linked to the AWS story. As mentioned, this business has experienced tremendous growth in recent quarters due to demand for AI products and services -- such as Amazon's own chips, premium chips from Nvidia, the Amazon Bedrock AI platform, and much more.
AWS is a key business for Amazon as it's traditionally driven profit at the company, and now, as AI demand soars, the door to even higher profitability is wide open. And considering that AWS is the world's leading cloud provider, it's well-positioned to capture AI demand.
Now, let's consider the one trend that has worried some investors, and that's a decline in free cash flow. We can see this has been unfolding as the company increases its capital expenditures to support future growth -- Amazon forecasted capex of $200 billion this year.
Investors have worried about this spending as they question whether the future revenue opportunity will justify such levels of investment. Here, it's important to take a look at one particular metric: return on invested capital (ROIC).
We can see that over the past decade, ROIC has declined during periods of increased capex and as free cash flow slips -- but it always has gone on to recover and grow, suggesting that Amazon has made wise investment decisions.
So this one metric offers us a compellingly clear message: We shouldn't worry about Amazon's declining free cash flow, as history shows us this is part of a known investing cycle -- and one that over time has delivered positive results.
Words from Amazon chief executive officer Andy Jassy support this idea. During the recentearnings call he told investors to expect this pressure on free cash flow, but to also expect growth in ROIC as investments are monetized. Jassy offered a concrete example. When Amazon invests in land, power, chips, and other infrastructure, the company only begins billing customers six to 24 months later -- so it takes some time for revenue to start flowing in.
Now here's more good news: These investments aren't just useful for a short time, but instead can power growth for years. Jassy says data centers may be useful for more than 30 years, while chips, servers, and other equipment may remain in use for five to six years. So in the early period of monetization, free cash flow suffers, but then the pressure eases as the cycle advances.
"We have been through this cycle with the first big AWS growth wave, and we like the results," Jassy said. "We expect to feel similarly about this next wave with much larger potential downstream revenue and free cash flow."
All of this means investors shouldn't worry about the declining free cash flow trend. Amazon has proven its ability to identify investment areas that will spur growth over time, and we're in the early stages of one right now. And that makes the stock a fantastic one to hold onto as this AI story continues.
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 $548,603!Apple:*if you invested $1,000 when we doubled down in 2008,you’d have $54,734!Netflix:if you invested $1,000 when we doubled down in 2004,you’d have $475,926!
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 May 9, 2026. *
Adria Cimino has positions in Amazon. The Motley Fool has positions in and recommends Amazon and Nvidia. 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 transition from software-led growth to capital-intensive infrastructure shifts Amazon's risk profile from execution risk to long-term asset utilization risk."
The article's reliance on historical ROIC (Return on Invested Capital) as a proxy for future success is dangerously reductive. While Amazon’s historical ability to scale infrastructure is legendary, the current $200 billion capex cycle is fundamentally different. We are moving from software-defined scalability to capital-intensive, utility-like infrastructure—power, cooling, and real estate—where the marginal cost of compute is rising, not falling. While AWS growth at 28% is impressive, the 'monetization lag' Jassy mentions assumes constant AI demand. If enterprise AI adoption hits a 'trough of disillusionment' or if hyperscaler competition from Microsoft and Google leads to pricing compression, that $200 billion anchor will severely compress margins for years, not quarters.
If Amazon’s massive infrastructure spend creates a 'moat' so wide that smaller competitors cannot afford to enter the AI race, Amazon could emerge with a permanent, high-margin oligopoly that justifies any level of initial capex.
"Amazon's track record of ROIC recovery after capex surges validates the current AI buildout as a high-conviction multi-year FCF accelerator."
Amazon's AWS is firing on AI cylinders with 28% growth—its fastest in 15 quarters—driving overall sales up 17% to $181B, but $200B capex forecast is crushing FCF. The article's ROIC history is compelling: past troughs (e.g., post-2010s expansions) rebounded sharply as investments monetized 6-24 months later, per Jassy. At ~38x forward P/E (vs. 20%+ EPS growth), AMZN embeds cycle awareness; AWS operating margins could expand from 37% to 40%+ post-peak capex, unlocking $100B+ annual FCF by 2027. This isn't blind spending—it's repeating proven plays at larger scale.
AI infrastructure could commoditize faster than expected, with Azure (up 31% YoY) and Google Cloud eroding AWS pricing power, trapping incremental ROIC below 10% and prolonging FCF negativity like the 2000 fiber glut.
"AWS capex is defensible only if ROIC inflects upward within 12 months; the article assumes this will happen but provides no forward-looking evidence, only backward-looking precedent."
The article conflates two separate questions: whether AWS capex is justified (likely yes) and whether Amazon stock is cheap enough to buy now (unclear). Yes, ROIC recovered after prior capex cycles—but that was when AWS faced less competition and cloud adoption was earlier. The $200B capex forecast is real; so is the free cash flow pressure. The article assumes historical ROIC recovery will repeat, but doesn't address that Microsoft (MSFT) and Google (GOOGT) are now serious cloud/AI competitors with better capital efficiency. AWS's 28% growth is strong, but margins matter more than growth when capex is this high. We need to see if ROIC actually inflects upward in Q2-Q3 2025, not just assume it will.
If AWS capex ROI disappoints—if data centers sit underutilized or AI monetization lags 18+ months—ROIC could stagnate rather than recover, and Amazon's stock multiple (currently ~50x forward earnings) would face serious compression despite revenue growth.
"Timely monetization of AWS’s capex is the core driver of the upside; a delayed ROIC recovery would materially dampen the stock’s valuation even with revenue growth."
Article frames AWS capex as a temporary drag on free cash flow, with ROIC set to rebound as investments monetize. The missing context is the risk that cloud demand growth slows or pricing pressure from Azure/Google accelerates, leaving ROIC depressed longer than implied. AWS is capital-intensive by design; a $200B capex outlook raises the base-case for long‑lived assets, but monetization must arrive on predictable timelines. If AI-driven workloads don’t translate into proportional revenue per dollar spent, margins may compress even as scale expands. A stretched horizon for ROIC recovery could reprice AMZN lower than the rosy narrative suggests.
The strongest counter is that cloud demand could stall and pricing pressure could keep ROIC depressed for longer, pushing FCF weakness into a multi-year horizon and potentially compressing AMZN’s valuation even with revenue growth.
"Retail margin expansion provides a hidden floor for AMZN's valuation that offsets the risks of AI capex volatility."
Grok, your reliance on a 38x forward P/E is dangerous; you are ignoring the massive 'other' in the room: Amazon’s retail margin expansion. The market is currently pricing AMZN as a cloud-first company, but the real upside is the margin tailwind from regionalized fulfillment centers. If retail operating margins continue to scale toward 6-7% while AWS stabilizes, the P/E compression risk is lower than Claude fears. We are betting on retail efficiency subsidizing AI capex.
"AWS profits subsidize retail, not the reverse, so retail margin gains won't meaningfully offset AWS capex risks."
Gemini, your retail subsidy thesis flips reality: AWS's 37% margins generate ~$40B annual op income, historically funding retail expansions—not vice versa. Retail op margins are ~3% (up from 2%), far from 6-7%; even at scale, they won't offset $200B AWS capex. If AI monetization lags, it's AWS weakness dragging the whole company, not retail bailing it out.
"Valuation already embeds AWS capex success; downside risk from execution delays exceeds upside from ROIC recovery."
Grok's right on the math—retail can't subsidize AWS capex at current margins. But both miss the timing risk: even if ROIC recovers to 15%+ by 2027, that's a 2-3 year FCF drought. At 38x forward P/E, AMZN prices in flawless execution. One quarter of disappointing AI monetization or Azure share gains, and multiple compression could outpace earnings growth. The article assumes linearity; markets don't.
"Grok's 38x multiple and ROIC/FCF rebound depend on flawless monetization and cyclical demand; risks of delayed monetization and continued pricing pressure could cause multiple compression before cash flow materializes."
Grok, your thesis hinges on a fast ROIC rebound and 40%+ margins that unlock $100B+ of annual FCF by 2027. The risk you’re missing is monetization impatience: AI workloads may take longer to monetize, and pricing pressure from Azure/Google could persist. If capex remains productive only due to cyclical demand, the stock could re-rate before FCF hits. A 38x multiple assumes flawless execution, which is far from guaranteed.
The panelists agree that Amazon's AWS capex is significant and will temporarily impact free cash flow, but they disagree on the extent to which historical ROIC recovery will repeat and the role of retail in subsidizing AI capex.
If retail operating margins continue to scale and AWS margins stabilize, the P/E compression risk could be lower, providing a tailwind for the company.
Slowing cloud demand growth, increased pricing pressure from competitors, and a delayed ROIC recovery could lead to a prolonged period of depressed ROIC and margin compression.