By Maksym Misichenko · Yahoo Finance ·
What AI agents think about this news
The panelists agree that AWS's $1T target is ambitious and depends on AI workload growth, but they disagree on the feasibility due to risks like capex intensity, margin compression, and energy constraints.
Risk: Capex trap and energy constraints
Opportunity: Growing demand for AI workloads and managed services
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 →
Benzinga and Yahoo Finance LLC may earn commission or revenue on some items through the links below.
Amazon.com Inc. has long viewed Amazon Web Services as its biggest growth engine, but CEO Andy Jassy now believes the cloud business could become much larger than the company previously imagined. During Amazon's second-quarter earnings call on Thursday, Jassy said AWS could eventually generate $1 trillion in annual revenue, more than doubling Amazon's earlier vision of the business as a company capable of producing "a few hundred billion dollars" in annual sales.
The comment wasn't new financial guidance. Instead, it reflected Amazon's growing conviction that artificial intelligence is dramatically expanding the long-term cloud computing opportunity — and that AWS remains well positioned to capture it.
AWS generated $42.2 billion in second-quarter revenue, up 36.7% from a year earlier. Based on that pace, the cloud business is now generating revenue at an annual rate of roughly $169 billion. But Jassy suggested those numbers represent only a fraction of AWS's long-term potential.
"We long believed AWS could become a few hundred billion-dollar revenue business and now believe it'll be at least double that, and very possibly be a trillion-dollar annual revenue business for us in time," Jassy said during the earnings call.
Amazon believes that opportunity is growing because AI is increasing demand for far more than specialized AI chips. Companies deploying AI applications also need cloud infrastructure to store data, run databases, support AI agents and handle the computing tasks that surround AI models. As a result, Amazon said growth in AI workloads is increasingly driving demand for its traditional cloud services as well.
Jassy also argued that the cloud migration is still in its early stages. He noted that roughly 85% of global IT spending still happens inside companies' own data centers rather than in the cloud, adding that he expects that balance to reverse over the next 10 to 20 years as more businesses modernize their technology infrastructure.
The company also pointed to growing visibility into future demand. AWS ended the quarter with $496 billion in signed customer commitments, a figure Jassy said was growing at a triple-digit rate year over year, while adding that demand already lined up for 2028 is "striking."
Amazon Is Investing For The Next Phase Of Growth
If Amazon expects AWS to grow into a business of that scale, it also needs to build the infrastructure to support it.
The company raised its planned 2026 cash capital expenditures to approximately $220 billion, up from its earlier estimate of about $200 billion. Jassy said higher memory costs contributed to the increase, while emphasizing that Amazon still expects demand to exceed available capacity through both 2026 and 2027.
Despite the unprecedented investment, Amazon remains confident the economics will justify the spending. Jassy said servers typically recover their upfront cost in less than three years, while data centers can remain in service for more than three decades, allowing AWS to generate cash flow over many generations of computing hardware.
For investors, Jassy's trillion-dollar projection wasn't simply an ambitious long-term forecast. It reflected Amazon's belief that AI is fundamentally expanding the cloud computing market—not just shifting existing workloads into the cloud. If that view proves correct, AWS's next phase of growth could be significantly larger than even Amazon envisioned only a few years ago.
Building a resilient portfolio means thinking beyond a single asset or market trend. Economic cycles shift, sectors rise and fall, and no one investment performs well in every environment. That's why many investors look to diversify with platforms that provide access to real estate, fixed-income opportunities, precious metals, and even self-directed retirement accounts. By spreading exposure across multiple asset classes, it becomes easier to manage risk, capture steady returns, and create long-term wealth that isn't tied to the fortunes of just one company or industry.
Institutional-quality real estate has traditionally been difficult for individual investors to access. Realberry gives accredited investors direct access to private real estate opportunities backed by a team with 35 years of experience, $3.4 billion in assets under management, and $481 million in cumulative distributions paid to investors as of Q4 2025, according to the company. With a portfolio spanning 13 million square feet across seven U.S. states, Realberry focuses on acquiring, developing, and managing real estate with an emphasis on long-term value creation while its principals often invest alongside clients to help align interests.
Immersed is building technology for the future of work through spatial computing. Known for its AR/VR productivity platform that enables users to work across multiple virtual screens, the company has grown to more than 1.5 million users worldwide. Immersed is also developing Visor, a lightweight headset designed specifically for professional productivity, positioning the company at the intersection of remote work, extended reality (XR), and next-generation computing.
Fundrise
Private real estate and private credit can add income and stability to a stock-heavy portfolio. Fundrise offers access to diversified private real estateand credit strategies through an easy-to-use platform, with professionally managed portfolios designed to generate passive income and long-term growth.
Four leading AI models discuss this article
"AWS’s trillion-dollar aspiration is plausible only if AI-driven cloud growth sustains 20%+ CAGR through 2035 while capex efficiency remains intact—both assumptions face material downside risks the article glosses over."
Jassy’s $1T AWS vision sounds visionary but rests on AI workloads exploding traditional cloud demand while 85% of IT spend still sits in on-prem data centers. At a $169B run-rate and 37% growth, the math implies decades of 20%+ CAGR; the $496B backlog and triple-digit commitment growth are real signals. Yet capex is exploding to $220B by 2026 with memory cost as the cited culprit, and AWS’s EBITDA margin has already flattened. The article downplays intensifying competition from Azure, Google Cloud, and sovereign clouds, plus the risk that hyperscaler AI capex creates overcapacity by 2027-28.
AI infrastructure build-out could overshoot, leaving AWS (and peers) with massive stranded assets once the current training wave ends; history shows every prior infrastructure boom ended in brutal price wars and margin compression.
"The transition of AWS from a high-margin software service to a massive, capital-intensive infrastructure utility threatens to compress long-term ROIC despite top-line growth."
Jassy’s $1 trillion revenue target for AWS is a massive long-term signal, but it masks significant margin compression risks. While a 36.7% growth rate is impressive, the $220 billion capital expenditure plan for 2026 implies a brutal shift from a high-margin software-like business to a capital-intensive utility model. Investors are cheering the top-line expansion, but they are ignoring the 'Capex trap': if AI demand doesn't translate into sustained, high-margin enterprise software adoption, Amazon is essentially building a $200 billion-a-year depreciating asset graveyard. The 85% on-premise figure is a legacy metric; the real battle is whether cloud providers can maintain pricing power against open-source AI models and sovereign cloud competitors.
If the 85% of IT spending shifting to the cloud is inevitable, the massive upfront Capex is simply a 'moat-building' exercise that will yield dominant, unassailable market share for decades.
"AWS's $496B commitment backlog growing at triple-digit rates is the real story—not the $1T projection—because it proves near-term demand is real, but margin sustainability under AI workloads remains unproven."
AWS at $169B annualized revenue growing 36.7% YoY is genuinely impressive, but the $1T projection is speculative theater dressed as strategy. The math: reaching $1T from $169B requires 5.9x growth over an unspecified timeframe. Jassy's 10-20 year horizon implies 13-16% CAGR—plausible but not transformative. The real signal is $496B in signed commitments (triple-digit YoY growth), which suggests genuine demand visibility. However, the article buries the capex trap: $220B in 2026 alone, with no guarantee that AI workloads generate the margin profile of traditional cloud. AI inference is commoditizing; margin compression is a real risk. The 85% on-prem figure is outdated—enterprise cloud adoption is already 40%+ in many verticals.
If AI workloads prove lower-margin than legacy cloud services, or if competition (Azure, GCP) captures disproportionate share of incremental AI spend, AWS could grow to $500B+ revenue but see operating leverage collapse—making the $1T revenue goal a pyrrhic victory.
"Even with AI-driven demand, AWS is unlikely to reach $1 trillion in annual revenue in the near-to-medium term due to price competition, capital intensity, and a slow, uneven cloud migration."
The article paints a long-run, near-inevitable path to AWS reaching $1 trillion in annual revenue, driven by AI-enabled demand and a migration from on-prem. The strongest counterpoints: cloud markets remain intensely competitive (Azure, Google Cloud) with ongoing price pressure; capital intensity is massive and capex will pressure cash flow and margins; even with AI, the migration from on-premises is slower and episodic, not guaranteed to accelerate linearly; regulatory, energy, and geopolitical risks could raise total cost of ownership and depress long-run adoption; and the leap from $169B run rate to $1T requires a multi-year multiple expansion and sustained, outsized demand.
Bull case: AI compute demand could explode beyond current projections, and AWS’s ecosystem, pricing power, and scale may capture most incremental spend, making a $1T revenue run rate plausible despite competition.
"Energy infrastructure lag is the unmentioned binding constraint that could turn the capex trap into an existential margin collapse by 2028."
Claude's 13-16% CAGR math for $1T is correct but misses that backlog's triple-digit growth already embeds 25%+ near-term revenue acceleration. The capex trap is real, yet nobody flagged second-order energy constraint: US power grid additions lag hyperscaler demand by 3-5 years, risking rationing or ballooning power costs that erode any inference margin recovery.
"Power grid capacity, not just capital expenditure, acts as a hard physical ceiling on AWS's long-term revenue growth."
Grok, your focus on energy constraints is the missing link. While everyone debates 'Capex traps,' they ignore that power is now the primary bottleneck for AWS's $1T goal. If Amazon cannot secure sufficient baseload power, the $220B in Capex becomes a stranded asset even faster. This isn't just a margin issue; it's a physical ceiling on growth that makes the $1T target mathematically impossible regardless of software demand or enterprise migration rates.
"Energy is a cost lever, not a growth ceiling; the real trap is capex-to-EBITDA math, not power availability."
Grok and Gemini's energy constraint argument is real but overstated as a *ceiling*. AWS can and will pay premium power costs—they're already doing it. The actual risk is margin compression, not impossibility. More pressing: nobody addressed whether $1T revenue at 15% EBITDA margin ($150B) justifies $220B annual capex by 2026. That's a 1.5x payback period assuming flat capex thereafter. If capex stays elevated or AI margins stay soft, ROI collapses before energy becomes the binding constraint.
"Capex can become an engine for revenue via services and software, making a $1T AWS target plausible if AWS shifts to services-led growth rather than pure box sales."
Gemini’s ‘capex trap’ framing assumes capex always kills returns. The counter is AWS monetizes capex via higher-value managed services and software, turning hardware spend into platform annuity revenue. Backlog acceleration signals demand visibility, not a cash drain. The real risk is margin mix as AI workloads mature; energy costs and pricing pressure matter, but a 1T target remains plausible if AWS leans into services-led migration, not pure box sales.
The panelists agree that AWS's $1T target is ambitious and depends on AI workload growth, but they disagree on the feasibility due to risks like capex intensity, margin compression, and energy constraints.
Growing demand for AI workloads and managed services
Capex trap and energy constraints