Superpower Daily
Home Depot Puts Magic Apron in All U.S. Stores, Adding Project...Chatbots Debunked Foreign Falsehoods About 75% of the Time, Bea...Blue Owl Leads $2.4 Billion IREN GPU Financing Tied to Hardware...CME Targets Oct. 5 Nvidia GPU Futures Launch as CFTC Tests the...Anthropic’s Claude Code Limit Dates Conflict, Obscuring a Repor...LAION Releases 10-Million-Hour Video Corpus for Research, Exclu...Home Depot Puts Magic Apron in All U.S. Stores, Adding Project...Chatbots Debunked Foreign Falsehoods About 75% of the Time, Bea...Blue Owl Leads $2.4 Billion IREN GPU Financing Tied to Hardware...CME Targets Oct. 5 Nvidia GPU Futures Launch as CFTC Tests the...Anthropic’s Claude Code Limit Dates Conflict, Obscuring a Repor...LAION Releases 10-Million-Hour Video Corpus for Research, Exclu...
Businesspublished

Barclays Sees Big Three Clouds Taking $35–$40 per $100 of AI-Lab Revenue

The bank’s model points to a lucrative serving layer for AWS, Azure and Google Cloud, even as falling training costs improve AI labs’ economics and self-built capacity looms.

By 2 min read
Barclays Sees Big Three Clouds Taking $35–$40 per $100 of AI-Lab Revenue
Barclays Sees Big Three Clouds Taking $35–$40 per $100 of AI-Lab Revenue
Sign in to save and react

Listen to this story

The audio brief

About 1:34
0:001:34
Read transcript
For every hundred dollars AI model companies bring in, Barclays estimates that AWS, Microsoft Azure, and Google Cloud collect thirty-five to forty dollars in inference-compute fees—and keep ten to twenty dollars as operating profit. That makes the cloud serving layer a major toll collector, even as the economics of AI labs improve. Barclays forecasts total AI-lab revenue rising from seven billion dollars in 2024 to 137 billion in 2026, then to 690 billion in 2028. At the same time, training is expected to shrink from 96 percent of lab revenue in 2024 to 35 percent in 2027 and 30 percent in 2028. As more spending shifts toward serving customers, paid inference margins could reach 50 to 65 percent or higher in 2026, driven mainly by enterprise customers and agentic workflows. The mix matters. Barclays estimates direct API inference margins above 80 percent, compared with roughly 70 percent for subscriptions. In its example, an API-heavy lab reaches about a 55 percent adjusted gross margin, versus 38 percent for a subscription-heavy one. Revenue-recognition choices can make that reported gap look even wider. For the next two years, Barclays expects the three cloud providers to hold their shares of lab-compute spending. The constraint arrives from 2028: secured, self-built AI infrastructure could reduce cloud spending on both training and inference. The key question is how quickly that owned capacity comes online.

Story brief

3 key points

Barclays projects AI-lab revenue will expand from $7 billion in 2024 to $690 billion in 2028, while cloud infrastructure remains a major toll collector. AWS, Azure, and Google Cloud are expected to retain their current shares of lab compute spending through 2027, earning roughly 35%–45% operating margins on inference-related fees. Labs’ economics should improve as training costs decline, but future self-built...

  1. 01

    Barclays forecasts training will fall from 96% of AI-lab revenue in 2024 to 30% by 2028.

  2. 02

    Paid inference margins could reach 50%–65% or higher in 2026, led by enterprise and agentic workloads.

  3. 03

    Direct API inference margins exceed 80%, versus roughly 70% for subscriptions, according to Barclays.

AI model companies may be improving the economics of serving customers, but the largest cloud platforms still sit deep in the revenue stream. Barclays estimates AWS, Microsoft Azure and Google Cloud receive $35 to $40 in inference-compute fees for every $100 AI labs generate—and retain $10 to $20 in operating profit.

A growing pool, with less spent on training

Barclays projects total AI-lab revenue will rise from $7 billion in 2024 to $137 billion in 2026 and $690 billion in 2028. Those are forecasts, but their scale makes the allocation of revenue between model companies and infrastructure suppliers commercially significant.

The bank expects training expenditure to fall from 96% of AI-lab revenue in 2024 to 35% in 2027 and 30% in 2028. Barclays expects labs’ overall profitability to improve as inference profit overtakes training expenditure, but serving those workloads would continue to send substantial spending to the cloud platforms.

Product mix changes what a margin means

The cloud figures do not translate into a single uniform lab margin. Barclays estimates paid inference margins rose from low double digits in 2025 to 50%–65% or higher in 2026, driven primarily by enterprise customers and agentic workflows.

Three reasons comparisons can diverge

  • Barclays estimates direct API inference margins exceed 80%, while subscription products carry margins of about 70%.
  • Its hypothetical API-heavy lab posts an adjusted gross margin of about 55%, versus about 38% for a subscription-heavy lab—a 17-percentage-point gap.
  • Revenue-recognition choices can widen the reported difference when indirect API sales are recorded on a gross basis, net basis, or not recognized by a lab.

Owned capacity is the later constraint

Barclays expects AWS, Azure and Google Cloud to hold their respective shares of AI-lab compute spending over the next two years. From 2028, it expects secured self-built AI infrastructure projects to come online and potentially reduce the providers’ share of both training and inference spending. The potential challenge is therefore a forecast tied to future capacity, not a market-share shift already recorded.

Sources

  1. finance.biggo.comFor Every $100 AI Model Companies Earn, $35–$40 Flows to the Big Three Cloud Providers — BigGo Finance

Related coverage

Continue reading