AI monetisation: 4.5% of US users pay, top payers spend $900/mo

What happened

Venture capital firm Andreessen Horowitz has, for the first time, included actual consumer spending data in its closely watched Top 100 AI list, offering a rare empirical look at how US consumers use and pay for artificial intelligence tools. According to the firm's analysis, nearly half of US consumers now use AI in some form, yet only 4.5 percent pay for a subscription.

Among that small paying minority, spending is sharply concentrated: the top one percent of payers spend roughly $900 a month on AI tools, with that outlay going predominantly toward professional-grade products such as development environments and automation tools rather than general consumer chatbots.

Why it matters

The data gives operators and investors a clearer picture of where real monetisation is happening in AI, cutting through usage-growth headlines that often conflate free trial activity with sustained paid demand. It suggests the AI market is bifurcating: a broad base of casual, largely free usage, and a much smaller, high-value segment of professional or power users willing to pay significant sums for tools embedded in their workflow.

For businesses building or buying AI products, this points to a need to rethink pricing and positioning. Mass-market free tiers may be effective for adoption and brand reach, but the commercial upside looks concentrated among specialists — developers, automation-heavy teams, and other professional users — who value AI as a productivity multiplier rather than a novelty.

By the numbers

• Nearly 50% of US consumers report using AI tools in some capacity.
• 4.5% of US consumers currently pay for an AI subscription.
• Top 1% of paying users spend an average of roughly $900 a month on AI tools.

The Renascence take

The headline gap between usage and payment is less surprising than what it implies about value perception. Most people experience AI as a convenience layered onto existing habits — search, writing, quick answers — which rarely justifies a subscription. The minority who pay heavily are doing something qualitatively different: embedding AI into a professional process where time saved has a clear monetary value.

This is a textbook case of value-based pricing hiding in plain sight. The lesson isn't "charge more for AI" — it's that organisations should stop pricing AI like a feature and start pricing it like a role it replaces or augments. The real opportunity lies in identifying the narrow set of high-intensity use cases where AI measurably compresses time or cost, and designing the experience — and the price tag — around that outcome, rather than chasing broad but shallow adoption.