Enterprise AI Software Is Finally Becoming a Revenue Line Investors Can Measure

For several years, investors had to infer enterprise AI demand from cloud growth, product launches and management commentary. That is changing.

A small group of large software companies now disclose AI-linked contract values, recurring revenue or usage metrics substantial enough to track across quarters.

ServiceNow crossed a useful threshold

ServiceNow said its AI business crossed $1 billion in annual contract value in Q2 2026. That figure is not the same thing as recognised revenue, but it gives investors a contractual measure of customer demand.

The significance is that AI is being sold into existing enterprise workflows rather than only tested as a separate product.

Salesforce is making the agent business more measurable

Salesforce reported Agentforce ARR above $1.5 billion in its second fiscal quarter, with Agentforce and Data 360 ARR near $3.9 billion. It also disclosed billions of agentic work units processed across its platforms.

Those measures still require interpretation because product definitions can evolve, but they are materially more useful than counting customer logos or demonstrations.

Microsoft shows the size of the platform layer

Microsoft reported $59.3 billion in quarterly cloud revenue and Azure growth of 43% in fiscal Q4 2026. Its commercial remaining performance obligation reached $678 billion.

Not all of that demand is AI. The point is that AI is being absorbed into an enormous existing commercial cloud and productivity base.

The next question is economic quality

Revenue growth does not settle the investment case. AI features can require expensive infrastructure, sales incentives and implementation support. Investors should track gross margin, free cash flow, renewal and expansion alongside AI-linked revenue.

The category is becoming easier to measure. That will also make weak monetisation harder to hide.