Jassy's barbell

On the latest Q2 call, Amazon’s CEO described AI demand as “very barbelled.”

• One end: "the AI labs are consuming gobs and gobs of compute, and there are a few runaway successful generative AI applications like Claude Code and ChatGPT."

• The other end: "enterprises who are getting real value from AI in cost avoidance and productivity… automating customer service or business process automation or fraud."

• The middle: "all of the current enterprise production workloads, some of which are using inference in a pervasive way, but most of which aren't."

​Then the line that matters for anyone selling into that middle:

"In my opinion, that will be the largest absolute segment, the existing production workloads in the enterprise and new businesses and workloads that startups build too."
Note: "I don't know if the trajectory of that middle part of the barbell will be the same wildly steep trajectory that we've seen with the current barbell AI labs piece."

Largest, but not fastest.

McKinsey just measured the same barbell from the buyer's side

Their State of AI survey, published August 25, covered 1,719 leaders.

• 80% say AI improved their individual productivity

• 37% attribute at least some EBIT impact to AI, essentially unchanged from 2025

• Only 6% meet McKinsey’s high-performer threshold: at least 5% EBIT impact plus significant value from AI.

Deployment is broadening. Material enterprise impact remains concentrated.

The high performers also widen the objective beyond efficiency. Most pursue growth or innovation, and ~3/4 fundamentally redesign workflows. Only 1/4 of other respondents do the same.

The next phase is operational: moving AI from focused use cases into existing workflows, applications and customer experiences.​

Why the middle is a Marketplace question

Production AI creates demand for cloud consumption, AI-native applications, implementation, security, governance and integration.

The budget for it is already contracted.

AWS ended Q2 with $496B in committed customer spend, up $132B in a single quarter.

But demand does not turn into (partner) revenue on its own. Vendors still need to attach their solutions to cloud budgets, earn seller attention, navigate procurement.

Marketplace and co-sell form part of that commercial infrastructure to turn AI technology into usage and scale.

​For alliance and Cloud GTM leaders, the challenge is turning that shift into an operating model: choosing the right Marketplace strategy for your stage, aligning internal sellers and cloud teams, and moving customers from purchase into deployment, usage and expansion.​

That execution gap is what we work through in the Cloud GTM Leader course.

​The next five-week cohort starts September 29. It is designed for alliance and Cloud GTM leaders building repeatable growth across AWS, Microsoft Azure and Google Cloud marketplaces.

Inside the course:

• Marketplace strategy from listing to traction and scale

• Co-sell, cloud relationships and differentiation

• Internal buy-in, sales enablement, KPIs and scaling frameworks

If you want to turn Marketplace into a repeatable GTM engine, Cohort 16 is open.

​Join September Cohort