Amazon's AWS AI and custom chip units each pass $25 billion run rate in Q2 2026
In brief
- AWS AI and custom chip businesses each passed a $25 billion annualized run rate in Q2 2026.
- AWS revenue reached $42.2 billion, up 37%, its fastest growth in 18 quarters.
- Amazon raised 2026 capex guidance to roughly $220 billion from $200 billion.
- Trailing twelve-month free cash flow fell to a $7.6 billion outflow.
What the Q2 numbers showed
These are Q2 2026 figures from late July. They aren't new results.
The two $25 billion figures are separate, and Amazon didn't combine them into a single total. A run rate also isn't booked annual revenue (it takes current revenue and projects it across a full year).
Jassy's comparison was about the AWS AI business, the unit that reached a $25 billion run rate in roughly three years. He measured it against AWS's own early days.
CEO Andy Jassy told investors on the July 30 earnings call that the unit is growing “several hundred times faster” than AWS did in its early days.
The wider cloud business kept pace. AWS revenue reached $42.2 billion in the quarter, up 37% from a year earlier, which works out to a $169 billion annualized run rate and AWS's fastest growth in the last 18 quarters. AWS posted an operating margin of 39.4%. Amazon also reported a backlog of $496 billion that grew at a triple-digit rate year over year, and it said demand was exceeding its capacity for 2026 and 2027.
Bedrock, Amazon's managed AI service, had a standout quarter. Its Q2 usage surpassed all previous quarters combined.
The spending side
Amazon raised its 2026 capital expenditure guidance to approximately $220 billion, up from $200 billion. The company tied most of that to AI infrastructure and custom chips. Crypto Briefing noted that rising memory costs put further pressure on the budget.
The cash flow shows it. Amazon's trailing twelve-month free cash flow fell to an outflow of $7.6 billion, largely because of the heavier investment.
Trainium, Amazon's in-house AI chip, sits at the centre of the chip story. Amazon has major agreements with both Anthropic and OpenAI to use it.
Where Crypto Briefing sees risk
Crypto Briefing's analysis flagged two pressure points. The first is capacity: Amazon can't recognize revenue it can't serve, so the timing of data center and chip buildouts is critical. The second is concentration, since Anthropic and OpenAI are anchor tenants for Trainium and much of the custom chip momentum depends on those relationships.
The demand is there. By Amazon's own account, the capacity isn't yet.
Frequently asked questions
What does a $25 billion run rate mean for Amazon's AI business?
A run rate takes current revenue and projects it across a full year. It's a snapshot of momentum, not a booked annual total. Amazon said its AWS AI business and its custom chip business each passed the $25 billion mark in Q2 2026.
Why did Amazon's free cash flow turn negative?
Amazon's trailing twelve-month free cash flow fell to an outflow of $7.6 billion, largely due to heavier investment. The company raised its 2026 capex guidance to approximately $220 billion from $200 billion, tying most of it to AI infrastructure and custom chips.
What risks did Crypto Briefing flag in Amazon's AI growth?
Crypto Briefing said capacity constraints mean Amazon can't recognize revenue it can't serve, which makes the timing of data center and chip buildouts critical. It also noted that Anthropic and OpenAI are anchor tenants for Trainium, so much of the custom chip momentum depends on those relationships.
