By TrendyVest.com | June 25, 2026
Let's talk about a market that has lost the plot.
We're living in a moment where Intel — a company posting a loss of $0.73 per share — trades at a forward P/E of 128x. A moment where ARM Holdings commands a trailing price-to-earnings ratio above 486x, and its controlling shareholder — SoftBank, which owns 87% of the company — has watched its investment compound to a valuation north of $300 billion on a business generating roughly $4 billion in annual revenue. A moment where SpaceX — which just completed the largest IPO in history, raising $75 billion at a valuation approaching $1.8 trillion — now trades publicly on Nasdaq at a market cap that has briefly surpassed both Amazon and Microsoft, despite posting a GAAP net loss of nearly $5 billion in 2025.
And then there's Amazon.
Amazon — which the market has decided to punish.
Let that sink in for a moment.
What the Market Is Actually Doing Right Now
Before we talk about Amazon, let's be precise about the distortion. The market isn't being irrational across the board. It's being selectively irrational — rewarding story stocks with metrics that strain credulity while penalizing one of the most operationally dominant companies ever assembled because it temporarily can't show you the cash.
Intel's stock has been among the most aggressively re-rated in the semiconductor sector. The company lost money — EPS of -$0.73 for the most recent quarter — and trades at a forward P/E of 128x, implying investors have already priced in a turnaround that has been promised, missed, and re-promised for the better part of a decade.
ARM Holdings has a trailing P/E above 486x. SoftBank, which holds an 87% controlling stake, has seen the value of its investment grow to something that would make most sovereign wealth funds envious. The market is pricing ARM as though it will shortly become one of the most profitable technology companies in history.
SpaceX, which just made history by completing the largest IPO ever recorded, now trades on public markets. Investors finally have audited financials to work with — and what they reveal is a company generating real revenue and growth alongside a GAAP net loss of nearly $5 billion in 2025, with an xAI segment burning $7.72 billion in a single quarter. The market's enthusiasm has pushed SpaceX to a valuation that briefly exceeded $2 trillion — for a company with roughly $19 billion in trailing revenue.
Meanwhile, Amazon — the largest cloud provider on earth, the dominant force in global e-commerce, the world's most advanced logistics operator, a pioneer of deployed robotics at scale, a chip manufacturer with $50 billion in addressable revenue, and the company currently building a satellite internet constellation that competes directly with Starlink — trades at a discount to all of them.
Not a mild discount. A significant one.
The consensus price target from 63 analysts is $312. The stock is near $225. That's roughly 38% upside sitting on the table, in plain sight, on a company that just posted one of the most remarkable quarters in corporate history.
Let's Actually Look at What Amazon Is
Here's what we're talking about when we talk about Amazon. Not the mythology. The facts.
The Cloud Business
AWS grew 28% year-over-year in Q1 2026 — its fastest growth rate in 15 quarters — generating $37.6 billion in a single quarter. That's a $150 billion annual revenue run rate, accelerating. The AI infrastructure backlog — pre-committed, pre-sold demand — is at a level no hyperscaler has ever seen, combining existing contracts, Amazon's commitment of up to 5 gigawatts of Trainium capacity to Anthropic, and OpenAI's expanded commitment of over $100 billion to AWS infrastructure over eight years.
Andy Jassy put it plainly in his shareholder letter: "We're not investing approximately $200 billion in capex in 2026 on a hunch."
He's right. The demand is already there. The backlog ratio — committed future demand relative to current revenue — is at a level no hyperscaler has ever seen. The previous record at a prior cycle peak was 1.7x. Amazon's is now 2.4x, and growing.
The Chip Business No One Is Pricing In
This is perhaps the most overlooked story in all of technology right now.
Amazon's custom silicon division — Trainium AI accelerators, Graviton CPUs, and Nitro networking chips — crossed a $20 billion annual revenue run rate in Q1 2026 and is growing at triple-digit rates. Trainium2 offers a 30% price-performance advantage over comparable Nvidia GPUs. Trainium3, which began shipping in early 2026, improves on that by another 30–40%. Trainium3 is nearly fully subscribed. Reservations are already being taken for Trainium4 — a chip that doesn't ship for another 18 months.
CEO Andy Jassy has said that if Amazon's chip business were a standalone company — selling this year's output to AWS and third parties the way Nvidia sells to the world — the annual revenue run rate would be $50 billion.
To put that in perspective: ARM Holdings, with its roughly $4 billion in annual revenue, trades at a trailing P/E above 486x and a market cap north of $300 billion. Amazon's chip business — generating $20 billion internally and pointing toward $50 billion externally — is valued at approximately zero by the market. It's bundled into the AWS number and forgotten.
Meta recently committed to using tens of millions of Graviton cores to power the CPU-intensive workloads behind its agentic AI systems. OpenAI runs on Trainium. Anthropic runs on Trainium. Uber runs on Trainium. The third generation chip is sold out before it finished shipping.
Amazon is winning the AI chip war quietly. The market hasn't noticed.
The Robotics Empire
Here's a number that should stop you cold: Amazon has deployed over one million robots across its global fulfillment network. The robot-to-human ratio in its warehouses is approaching 1:1.
This isn't a pilot program or a press release. It's the largest deployed robotics operation in the history of commerce. Sequoia, Amazon's integrated automation platform, processes inventory up to 75% faster than human-only operations. The newest fulfillment centers use 10 times more robots than previous designs, reducing fulfillment costs and processing times by 25%.
Amazon is opening multiple 3-million-square-foot fully-automated fulfillment centers in 2026, with a goal to automate 75% of all fulfillment operations by 2027. The AI model coordinating all of this — DeepFleet — optimizes the movement of every robot across every facility in real time.
When the next wave of AI comes for physical logistics, Amazon doesn't have to adapt. It built the infrastructure. It owns the data. It runs the robots. It writes the software. Everyone else will be catching up to what Amazon already operates today.
The Satellite Business
Amazon Leo — formerly Project Kuiper — has 367 production satellites in orbit as of June 2026, making it the third-largest satellite constellation in the world. Amazon has secured 92 rocket launches across ULA, Arianespace, Blue Origin, and SpaceX's Falcon 9, for a total launch commitment exceeding $10 billion — the largest commercial launch procurement in history.
Now here's the comparison that should make every investor pause.
SpaceX — which completed its IPO on June 12, 2026 — now trades publicly, and investors have the financials to examine. The picture is impressive in some ways: Starlink passed 10.3 million subscribers in Q1 2026, and the connectivity business is growing rapidly. But SpaceX posted a GAAP net loss of nearly $5 billion in 2025. Its xAI segment burned $7.72 billion in a single quarter. And the market has priced all of this at a valuation that briefly exceeded $2 trillion.
Amazon is building a competing global satellite internet constellation — with a public balance sheet, a proven hardware manufacturing operation, an FCC-licensed spectrum position, and $10 billion in launch contracts already signed.
SpaceX, with its AI segment and public market fanfare, commands a valuation that dwarfs Amazon's. Amazon gets approximately zero credit for building the same satellite connectivity business — plus everything else it operates.
The AI Beneficiary Nobody Is Talking About
Every conversation about AI beneficiaries runs through Nvidia, Microsoft, Google, and Meta. Amazon appears as an afterthought.
This is backwards.
Think about what AI actually requires: compute infrastructure, data storage, custom silicon, a global logistics network to deliver physical AI products, a consumer platform with hundreds of millions of active users to deploy AI into, and the financial scale to fund the arms race. Amazon has every single one of these in greater abundance than nearly any other company on earth.
Rufus, Amazon's AI shopping assistant, is used by hundreds of millions of customers and adding incremental annualized sales at scale. Amazon Bedrock — its AI platform — serves over 125,000 enterprise customers. The advertising business, powered increasingly by AI-driven targeting, has cleared $70 billion in trailing revenue at 24% growth, at margins closer to software than retail. Amazon Leo will connect hundreds of millions of underserved people to the internet — which is also the Amazon commerce platform, the AWS ecosystem, and the Alexa AI layer.
If you had to design the perfect AI company from scratch — one that touches every layer from silicon to satellite, from warehouse robot to consumer interface — you would design Amazon.
So Why Is It Down?
Here it is. The entire bear case, stated plainly:
Amazon guided $200 billion in capital expenditure for 2026. That was $50 billion more than Wall Street expected. Free cash flow dropped sharply as a result. The stock fell sharply in the weeks after the announcement.
That's it. That's the whole thing.
The market looked at a company investing the largest sum in corporate history into infrastructure that has committed customer demand behind it — with backlog ratios never seen before in cloud computing — and decided the appropriate response was to sell off a significant portion of the company's value.
The bears will say: capex could run ahead of monetization. AWS margins compressed slightly. Long-term debt has climbed. Free cash flow could stay negative into 2027.
All of that is true. None of it changes what Amazon is building.
Here's the question the market refuses to ask: what does this infrastructure become when the spending normalizes?
The AWS backlog is enormous — pre-sold. Trainium3 is sold out. Trainium4 is already reserved. Amazon Leo is live with 367 satellites and growing. The robotics network is processing hundreds of millions of orders a month with a 1:1 robot-to-human ratio. The advertising business is compounding at 24%.
The Comparison That Should Keep You Up at Night
Let's put three companies side by side.
ARM Holdings: Market cap north of $300 billion. Roughly $4 billion in annual revenue. Trailing P/E above 486x. SoftBank controls 87% of the stock and has every incentive to keep the narrative alive.
Intel: Negative earnings — EPS of -$0.73 for the most recent quarter. A forward P/E of 128x. Trading on the hope that a foundry turnaround materializes after years of missed targets.
Amazon: Near $250 per share. $23.9 billion in quarterly operating income. AWS growing at 28%. A chip business worth $50 billion that the market values at zero. One million robots deployed. 367 satellites in orbit. A massive pre-sold backlog. A consensus analyst price target of $312, with 63 buy ratings and zero sells.
The market is rewarding narrative and punishing execution.
The Bottom Line
Amazon is not struggling. Amazon is building.
There is a difference — and right now, Wall Street has confused the two.
The $200 billion capex isn't a warning sign. It's the most well-documented, pre-committed, customer-validated infrastructure investment in the history of technology. Every dollar of it has a name attached. OpenAI. Anthropic. Meta. The backlog doesn't lie.
The free cash flow will return. It always does when the build phase ends. And when it does, the company standing on the other side of this capex cycle will be something the market hasn't yet had the imagination to price.
The largest cloud on earth. The most advanced logistics network ever built. A chip manufacturer competing with Nvidia. A satellite constellation competing with Starlink. An AI platform serving 125,000 enterprises. A consumer interface used by hundreds of millions of people. All of it connected, compounding, and accelerating.
And Wall Street is selling it because the cash flow statement looks messy this year.
That's not analysis. That's impatience dressed up as discipline.
The most mispriced stock on Wall Street isn't a startup burning through runway on a dream.
It's the most complete company on earth — and it's on sale.
TrendyVest.com | June 25, 2026. This article is for informational purposes only and does not constitute financial or investment advice. Always conduct your own research before making investment decisions.