The ugly economics of consumer AI — SkimNews
Get the Tech newsletter
Daily tech — startups, AI labs, chips, the launches that shape the next decade. Free.
- Meta's Muse, OpenAI's Dots, and Instinct are recent consumer AI standouts, with Instinct reaching a $10B valuation on agentic errand-running for travel bookings and subscription cancellations.
- PNC research, cited by Andreessen Horowitz, found that as of May only 2.2% of consumers paid for AI at an average of $31 per month, with growth tracking linearly despite major model improvements.
- Bank of America reported in March that roughly 3% of U.S. consumers paid for AI, up 40% year-over-year, while a Menlo survey in September found 25% of adults use AI daily and half of those users pay for it.
- Scaling consumer AI to Netflix levels — 325 million subscribers at $34 — would only generate about $11 billion in annual revenue, less than a third of OpenAI's operating costs.
- OpenAI's widely reported enterprise pivot has paid off: enterprise bookings reportedly doubled since July, and even the Dots launch highlighted software-engineer and creative-agency use cases.
- Frontier labs have shifted toward the "Anthropic model" of enterprise contracts and vertical-by-vertical expansion because AI's operating costs are unusually high compared with social networking or cloud computing, capping how large any consumer business can grow.
Why it matters: At $31 per month with only 2.2% paying users, even Netflix-level saturation would yield roughly $11B in annual revenue — under a third of OpenAI's operating costs. That structural math is why frontier labs have already pivoted to enterprise contracts, and why any new consumer AI hit still needs an enterprise angle to clear breakeven.
Ask SkimNews
