The Neuron Daily reported A16z finding that top 1% of AI payers account for 19.5% of total consumer AI spending
A16z: Top 1% of Consumer AI Payers Average $903/Month
Consumer AI spending is concentrating heavily at the top, with a small share of users driving a disproportionate share of revenue. This concentration creates a structural advantage for model labs over app developers when serving the heaviest users.
The full picture
Andreessen Horowitz research shows the top 1% of paying US consumer AI users now average $903 per month, up from $504 in January 2025, and account for 19.5% of all consumer AI spending, more than the bottom 50% of users combined. The median consumer AI spend has barely moved, remaining around $25 per month. One top-1% payer spends as much as roughly 59 bottom-half payers on average. Model labs are better positioned than API-dependent app developers to profit from these heavy users, because labs pay only serving costs while apps pay API prices. Anthropic said one $200/month Max subscriber ran up tens of thousands of dollars in usage, a cost no API-paying app could absorb. Apps targeting heavy users may need to own more of the model stack, as Cursor did with its Composer coding model. Perplexity CEO Aravind Srinivas, speaking separately on the same dynamics, said some Perplexity Computer users spend upwards of $10,000 a month running agent loops, and that certain engineers at Meta and other companies generate around $10 million a year per engineer in AI coding tool costs.
How it developed
Perplexity CEO Aravind Srinivas described $10,000/month Perplexity Computer users and ~$10M/year-per-engineer AI coding costs at Meta
Sources
Want this in your inbox?
I send one email each morning with the stories that moved. If you would rather just read here, that works too.
Subscribe free