Palantir Posted 93% Growth. Amazon Is Spending $220 Billion. The AI Trade Is Accelerating — Not Slowing.
Palantir Posted 93% Growth. Amazon Is Spending $220 Billion. The AI Trade Is Accelerating — Not Slowing.
The doubters had a thesis: AI spending was peaking, returns were vague, and the trade was running out of steam. This week's earnings just torched that narrative. The numbers are not subtle.
Palantir's Quarter Was, in Its Own Words, "Otherworldly"
Palantir (PLTR) reported Q2 2026 results on August 3rd that stopped analysts mid-sentence. Revenue hit $1.94 billion — up 93% year-over-year — blowing past the $1.81 billion consensus by $130 million. But the headline wasn't even the most important number.
U.S. commercial revenue — the segment that tells you whether real businesses are paying real money for AI software — surged 149% year-over-year to $764 million. The company closed 220 deals worth $1 million or more, a record. And it raised its full-year revenue guidance to $8.15–$8.16 billion, up sharply from its prior $7.65–$7.66 billion forecast.
The metric that separates elite software companies from everyone else — the "Rule of 40" score (revenue growth + profit margin) — hit 155. Anything above 40 is considered healthy. Above 100 is nearly unheard of. Palantir is running at 155.
The stock had fallen 29% year-to-date going into the print. The market had priced in a slowdown. The business delivered the opposite.
Amazon Raised Its AI Spending to $220 Billion — And Says It's Still Not Enough
If Palantir shows AI software demand is real, Amazon shows the infrastructure buildout has no ceiling in sight.
AWS — Amazon's cloud division — grew 37% year-over-year to $42.2 billion in Q2, its fastest growth rate since 2021, trouncing analyst expectations of 31%. Total Amazon revenue hit $200.6 billion, up 20% year-over-year.
The bigger story: Amazon CEO Andy Jassy raised the company's 2026 capital expenditure target to $220 billion, up from a prior estimate of $200 billion, driven in part by surging memory costs. His message to investors was blunt: even at $220 billion, Amazon will not have enough capacity to meet demand this year — or in 2027. The AWS contract backlog stood at $496 billion at quarter end, up from $364 billion the prior quarter.
There is a catch worth knowing. Amazon's free cash flow flipped negative — a $7.6 billion outflow in the trailing twelve months, compared to an $18.2 billion inflow a year ago. The AI buildout is real. So is the bill.
Across the four largest hyperscalers — Microsoft, Alphabet, Meta, and Amazon — combined capital expenditure is now expected to reach $760 billion in 2026, up from $413 billion in 2025.
The Macro Backdrop: Chips at Records, Jobs Softening, Fed on Hold
The semiconductor industry just keeps printing records. Global chip sales hit $403.3 billion in Q2 2026, up 35.1% from Q1. May alone posted $120.6 billion in monthly sales — a 104.1% year-over-year gain and the 15th consecutive monthly record.
AMD added its own signal at the Advancing AI 2026 event, launching the Instinct MI455X — a 2nm chip with 432GB of HBM4 memory — paired with its Helios rack-scale platform now in full production. AMD claims the Helios system delivers 30% more tokens per dollar than Nvidia's competing Rubin rack. Competition is heating up at the chip level, which is structurally good for buyers — the hyperscalers spending $760 billion.
On the macro side, the picture is more complicated. The Federal Reserve held rates steady on July 29th, with inflation still running above its 2% target. The labor market is softening: June added just 57,000 nonfarm payroll jobs, well below expectations, and ADP's July private payrolls came in at just 44,000. The official July jobs report dropped this morning (August 7th), with consensus calling for around 80,000 new jobs.
A slowing jobs market cuts two ways for AI investors. It raises pressure on the Fed to eventually cut rates — good for high-multiple tech stocks. But it also raises the harder question: is AI absorbing resources faster than it is creating broad economic returns?
For now, the earnings data says the AI trade is very much alive. The infrastructure is being built at historic speed. The software layer is monetizing faster than almost anyone expected. The macro is the wildcard to watch.
New to AI investing? MasicotAI tracks the intersection of AI and markets every week, in plain language. [Get Started Today].
---
MasicotAI — Tracking the intersection of artificial intelligence and economic reality.