Palantir reported second-quarter revenue of $1.94 billion, up 93% from a year earlier, and raised its full-year forecast to between $8.150 billion and $8.158 billion. The company’s U.S. commercial revenue rose 149% to $764 million, according to its Aug. 3 investor presentation.

The numbers are Palantir’s, and one exceptional quarter should not be mistaken for a complete picture of enterprise adoption. But they provide unusually strong evidence that at least one class of AI spending is advancing beyond experiments: software tied directly to operating data, permissions and decisions.

From access to execution

The first wave of enterprise generative AI often centered on giving employees access to a chatbot or coding assistant. Those tools can be valuable, but their business impact is difficult to separate from ordinary productivity improvement. Palantir sells a different proposition. Its Artificial Intelligence Platform is designed to place models inside governed workflows that can use company data and trigger actions.

That distinction helps explain why U.S. commercial growth matters more than the headline alone. A business can trial a general-purpose assistant with a limited budget and few process changes. Connecting AI to a supply chain, manufacturing operation or customer workflow requires data integration, security controls, clear ownership and a willingness to redesign work.

The company also reported $2.13 billion in U.S. commercial total contract value during the quarter, up 153% year over year. Contract value is not the same as recognized revenue, and customers can still expand more slowly than sales projections imply. Even so, it is a sign that buyers are making larger commitments around production use.

What the market is rewarding

Palantir’s performance does not prove that every enterprise is ready for deep AI integration. Research on S&P 500 disclosures still shows that most large companies have not reached that stage. It does suggest that buyers who are ready may concentrate spending on systems that can demonstrate operational results.

That creates a harder competitive standard for AI vendors. A strong model remains essential, but model quality alone is becoming less defensible as companies can choose among commercial and open alternatives. The more durable value may sit in the layer that understands how a company works: its data, rules, permissions, processes and measures of success.

The enterprise AI market is therefore beginning to split. One side sells intelligence as a broadly available capability. The other sells the organizational machinery required to put that intelligence to work. Palantir’s quarter is an argument that the second category is becoming a serious budget line.

Sources for editorial review

Drafting note: This draft was prepared with AI assistance from the linked source material and requires author review, independent fact-checking and final editorial approval before publication.