By Maksym Misichenko · Nasdaq ·
What AI agents think about this news
The panel's net takeaway is that Amazon's future is uncertain, with potential growth in AWS and AI offset by risks in market share erosion, margin pressure, and competition.
Risk: AWS market share erosion and margin compression due to intense competition and AI infrastructure costs.
Opportunity: Potential acceleration in AWS growth driven by AI and a massive backlog.
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 →
It has been about five years since Amazon (NASDAQ: AMZN) founder and CEO Jeff Bezos stepped down from his role as CEO and handed over the reins to Andy Jassy.
Those five years were pretty favorable for the stock market, but not so much for Amazon stock.
Where to invest $1,000 right now? Our analyst team just revealed what they believe are the 10 best stocks to buy right now, when you join Stock Advisor. See the stocks »
Amazon has posted an average annualized return of just 6.8% over the past few years. That not only trails all of the other "Magnificent Seven" stocks, it's also worse than the S&P 500 and the Nasdaq-100.
The S&P 500 averaged an 11.5% annualized return over that stretch, while the Nasdaq-100 averaged a 14.2% return. The only Magnificent Seven stock that came remotely close to Amazon's underperformance was Microsoft (NASDAQ: MSFT). The others all averaged double-digit percentage annualized returns.
Even this year, Amazon stock has lagged the S&P 500 -- the stock is up 7% year to date, while the index is up 9% -- but it is beating several of its magnificent brethren.
So why has Amazon stock underperformed?
The tech giant hasn't navigated the AI boom as well as some of its competitors. While Amazon Web Services (AWS) remains the world's largest cloud computing infrastructure provider, it has steadily lost market share to rivals Microsoft and Alphabet (NASDAQ: GOOG) (NASDAQ: GOOGL). In 2021, Amazon's cloud market share was about 33%, but now it's down to about 28%. Meanwhile, Microsoft and Alphabet's Google have gained market share.
At the same time, Amazon has made massive investments in AI data centers, but many investors don't see those investments paying off well enough.
In its e-commerce business, growth has slowed since the pandemic-era surge. Shoppers' shift back toward normal retail behaviors initially left Amazon with an excess of inventory and shrinking margins. In more recent years, rising inflation and costs have also cut into its margins.
So where does Amazon go from here? Will the next five years be better for Amazon and its investors than the last five?
Amazon stock is up 8% year to date and up 14% over the past 12 months, but it still lags the S&P 500 over both time frames.
However, its growth numbers have improved. Jassy has also repeatedly reminded investors that the company's heavy spending on AI infrastructure is needed to grow AWS. Last quarter, net sales for AWS increased by 28% to about $38 billion. That was higher than the 23% growth rate in the fourth quarter or its 19% pace in 2025.
In fact, on the first-quarter earnings call, Jassy highlighted that it was the best growth rate in 15 quarters. He also said the company has $364 billion in contracted backlog, not including its recently inked $100 billion deal with Anthropic. That's up from a backlog of just $244 billion in the previous quarter.
Amazon is also seeing significant revenue growth from its Trainium AI accelerator chips. Jassy said recently that these chips could be a $50 billion business for Amazon. It has already secured some $225 billion in revenue commitments for the chip business.
The years of underperformance have made Amazon stock less expensive. It trades now at 29 times forward earnings. At that valuation, with the backlog Amazon has amassed, its promising AI chip venture, and a data center build-out that should start to pay off, the stock looks like a solid buy.
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:
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 July 20, 2026. *
Dave Kovaleski has no position in any of the stocks mentioned. The Motley Fool has positions in and recommends Alphabet, Amazon, Apple, Meta Platforms, Microsoft, Nvidia, and Tesla. 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
"Amazon's AI infrastructure bets are real but unproven at scale; valuation reflects optimism that must still be earned against ongoing cloud share losses."
The article portrays Amazon's post-Bezos era as a temporary dip, citing accelerating AWS growth (28% last quarter), a $364B backlog plus $100B Anthropic deal, and $225B in AI chip commitments positioning Trainium as a future $50B business. At 29x forward earnings it looks cheaper than historical norms. However, the piece glosses over persistent AWS market-share erosion (33%→28%), margin pressure from retail and capex intensity, and the risk that hyperscaler AI spend normalizes before Amazon's infrastructure investments yield commensurate returns. Five-year underperformance versus the Mag7 and broader indices reflects genuine execution gaps that a single strong quarter does not erase.
If AWS growth reaccelerates to mid-20s while retail margins stay compressed and AI capex balloons further, Amazon could continue to lag the S&P 500 and Nasdaq-100 for another 2–3 years, making the current 29x multiple look expensive rather than attractive.
"Amazon’s accelerating AWS growth and massive contracted backlog suggest a significant earnings inflection that current forward valuations fail to fully price in."
The article’s fixation on five-year underperformance is a rearview mirror trap. It ignores the massive structural shift Andy Jassy executed: moving from a bloated, pandemic-era fulfillment model to a lean, high-margin advertising and cloud-first powerhouse. AWS growth accelerating to 28%—the best in 15 quarters—is the real signal, not the lagging share price. When you combine that cloud re-acceleration with the massive $364 billion backlog and the potential of custom silicon like Trainium, the 29x forward P/E looks like an attractive entry point. Amazon isn't just a retailer anymore; it's a high-margin infrastructure play hiding in plain sight.
The bear case is that Amazon’s retail margins remain fragile and vulnerable to Temu and Shein’s aggressive pricing, which could force Amazon to sustain higher fulfillment costs that cannibalize AWS profits.
"AMZN's valuation has compressed to fair value, not a discount, masking that AWS growth is decelerating and margin expansion depends entirely on capex payoff that remains 2-3 years away."
The article conflates two separate problems: AWS market share erosion (real, structural) and valuation compression (cyclical). At 29x forward P/E, AMZN isn't cheap—it's fairly valued for a 28% growth rate in AWS, which the article itself notes is decelerating from 2021 levels. The $364B backlog is impressive but contracts, not revenue; AWS margins matter more than gross bookings. The Trainium chip business ($225B in commitments) is speculative and faces entrenched competition from NVIDIA and custom silicon from hyperscalers. The article cherry-picks Q1 AWS growth (best in 15 quarters) while ignoring that AWS operating margin likely compressed due to data center capex. AMZN's e-commerce margin pressure is real and ongoing.
If AWS stabilizes at 25%+ CAGR and Trainium captures even 5% of the AI accelerator TAM, AMZN's forward multiple could re-rate to 35x+ within 18 months, making today's entry attractive. The backlog visibility is genuinely rare for infrastructure plays.
"AMZN trades at about 29x forward earnings with a thesis tied to AI and AWS backlog, but real upside requires sustained margin expansion and credible AI monetization, which are uncertain."
Amazon's underperformance is real, but this article leans on AI and backlog as near-term catalysts while downplaying the risks. The bullish thread is AWS's 28% Q1 growth to $38B and a $364B backlog, plus Trainium and a $100B Anthropic deal, which could drive a multi-year revenue and spending cycle. The stronger counterpoints: backlog is not guaranteed revenue, and AWS margins could stay pressured as capex runs hot and AI infrastructure costs scale. Competitive pressure from Microsoft and Alphabet could erode cloud-share gains, and e-commerce margin normalization remains fragile. At ~29x forward earnings, any miss on AI monetization or margin recovery could re-rate the stock.
The thesis rests on a long AI cycle and monetization that may disappoint; if AWS margins stall or cloud share recedes, the current multiple could compress quickly.
"AI utilization upside could offset margin pressure faster than consensus expects, a scenario still unpriced."
Claude correctly flags that backlog ≠ revenue and Trainium remains speculative, yet the panel underweights second-order effects: if Anthropic's $100B commitment accelerates AWS consumption while hyperscalers normalize capex, margin compression could prove temporary. Nobody has modeled the scenario where AI-driven utilization lifts AWS op margins back toward 35% even as retail stays challenged.
"AI-driven workload shifts will structurally compress AWS operating margins despite higher utilization rates."
Grok, your hypothesis on 35% AWS margins ignores the structural shift toward inference-heavy workloads. As AI scales, AWS faces a 'margin mix' problem: high-margin legacy storage and compute are being cannibalized by lower-margin, compute-intensive AI workloads. Even with high utilization, the cost-per-token economics are deflationary for the provider. I disagree with the notion that AI utilization naturally reverts margins to historical peaks; the competitive pricing pressure from Azure and GCP makes that outcome highly unlikely.
"AWS margin compression is real for commodity inference, but enterprise bundling with ads and legacy workloads may preserve 32–34% op margins, not collapse to 25%."
Gemini's inference-margin cannibalization thesis is sound, but it assumes AWS pricing power collapses uniformly. Reality: enterprise workloads (legacy + AI) bundle differently than hyperscaler self-serve. Amazon's advertising business ($55B+ run rate, ~40% margins) cross-sells cloud infrastructure at negotiated rates, not spot pricing. If enterprise AI adoption outpaces hyperscaler capex normalization, bundled pricing could sustain 32–34% AWS op margins even as inference commoditizes. The panel hasn't separated enterprise versus hyperscaler margin trajectories.
"AWS margins are unlikely to re-rate to 35% solely on AI utilization; price competition and capex normalization cap margin upside, increasing downside risk to the 29x multiple if AI monetization misses."
Grok's push on a potential 35% AWS operating margin, driven by AI utilization alongside a still-challenged retail backdrop, assumes a clean mix shift and pricing power that ignores competition and capex normalization. In reality, AI workloads commoditize inference, hyperscaler pricing pressure remains, and enterprise bookings may lag. If margins settle in the low-to-mid 30s instead of re-rating toward 35%, the 29x forward multiple already baked in growth is at greater risk of multiple compression than uplift.
The panel's net takeaway is that Amazon's future is uncertain, with potential growth in AWS and AI offset by risks in market share erosion, margin pressure, and competition.
Potential acceleration in AWS growth driven by AI and a massive backlog.
AWS market share erosion and margin compression due to intense competition and AI infrastructure costs.