OpenAI and Meta confront tough consumer AI economics

The recent launches of Meta's Muse assistant and OpenAI's Dots, alongside Instinct's rise to a $10 billion valuation, suggest a consumer AI resurgence. Instinct, which recently raised a $1 billion Series C, focuses on agentic tasks like booking travel and canceling subscriptions. However, these consumer-facing products face a harsh financial reality. Frontier labs are increasingly shifting toward Anthropic's enterprise-focused model because consumer willingness to pay for AI has hit a clear ceiling, even as underlying models make massive performance leaps.

Data from Andreessen Horowitz and PNC research shows that as of May, only 2.2 percent of consumers paid for AI services, spending an average of $31 monthly. This linear growth barely registered the massive performance jump from GPT-5.2 to Astra. Bank of America reported in March that roughly 3 percent of U.S. consumers paid for AI, representing a 40 percent year-over-year increase. Meanwhile, a September Menlo survey offered a more optimistic view, finding that 25 percent of adults use AI daily, with half of those users paying for the services.

The core issue is that AI is exceptionally expensive to run. Even if an AI service achieved Netflix-level saturation of 325 million subscribers paying $34 monthly, it would generate just $11 billion in annual revenue. This amount represents less than a third of OpenAI's operating costs. To survive, OpenAI has pivoted to business clients, doubling its enterprise bookings since July. Even its new Dots assistant features an enterprise angle targeting software engineers and creative agencies.

For AI practitioners and developers, these economics mean that building purely consumer-focused applications is highly risky without an enterprise monetization strategy. Meta can subsidize Muse using its personalized ad targeting, while Instinct plans to take a transaction cut on purchases and avoid training its own frontier models. For most developers, however, the path to profitability requires targeting business budgets. Designing tools that can easily transition from personal assistants to enterprise workflows is becoming the standard playbook for sustainable growth.