The artificial intelligence tools millions of people use every day - ChatGPT, Claude, Gemini and others - may be significantly underpriced, and that window could be closing.
A new analysis published Thursday by Axios reveals that major AI companies are operating at steep losses, effectively subsidizing consumer and business access to keep users hooked before the bill inevitably comes due.
OpenAI is projected to burn $14 billion in 2026, up from $8 to $9 billion in 2025. Fierce competition has pushed labs to price aggressively, squeezing profits. Even as models get cheaper and faster to run, total usage is surging - and the math still doesn't pencil out.
In February, 28% of corporate OpenAI chat spending flowed through personal consumer plans rather than higher-margin enterprise tiers, according to expense-tracking firm Ramp - meaning the cheapest, most heavily subsidized subscriptions are also among the most widely used.
The dynamic has a familiar ring to anyone who watched the tech industry's last wave of disruption. Silicon Valley has seen this movie before. The so-called "millennial lifestyle subsidy" meant venture capital helped underwrite cheap Uber rides and DoorDash deliveries. Before that, Amazon built its customer base with low prices, free shipping, and for years, no sales tax in most states. Eventually, all of those companies had to charge enough to cover costs - and turn a profit.
AI companies are widely expected to follow the same arc. Both OpenAI and Anthropic are widely expected to go public, and public investors will demand earnings growth and expanding margins. OpenAI's internal documents, first reported by The Information, project the company won't turn a profit until 2029 - by which point it expects to post cumulative losses of $44 billion.
Pricing experts say the free-access model was an inevitable early strategy to familiarize markets with poorly understood technology. One analyst put it bluntly: "This AI land-grab is on a colossal scale, and the price tag for dominating this new world is equally colossal. Monetizing the services and recouping some of that investment is going to force some pretty significant changes in business models and service pricing, and those changes are likely to happen fast."
There is a silver lining: the costs of AI are still falling. New chips are more efficient, and token pricing - the cost of generating text - has dropped significantly. But even as individual query costs decline, total infrastructure spending keeps rising to meet surging demand.
For now, everyday users and businesses are the beneficiaries of one of the most expensive customer acquisition campaigns in tech history. Whether that pricing holds depends on how quickly AI companies can find a sustainable path to profitability - or how patient their investors remain.