The AI industry has a narrative problem. On one hand, we're told that artificial intelligence will make human work obsolete. On the other, we're asked to pay $200 a month for the privilege of using it. Something doesn't add up. Every week there's a new headline about AI replacing jobs, a new think piece about universal basic income, a new tech CEO reassuring us that the future will be just fine. But when you look at the actual numbers, the business models, and the economics holding it all together, the story starts to fall apart.

The pitch vs. the price tag

The sales pitch is seductive: AI will do your work for you. You won't need to grind anymore. The robots are here, and they're going to set you free. But freedom apparently costs $20 a month at minimum, $200 if you want the good stuff. OpenAI's Pro tier runs at $200 per month for "virtually unlimited" access. Anthropic's Max plan costs the same. And even at those prices, rate limits keep getting tighter. ChatGPT Plus users are now capped at 160 messages every three hours, with thinking model access limited to 3,000 messages per week. OpenAI's own head of ChatGPT, Nick Turley, has said the company expects pricing to "significantly evolve," and that unlimited subscription plans may not last. The flat-rate pricing model that defined the early AI era is cracking under the weight of agentic usage, where tasks that once cost thousands of tokens per session now burn through millions. Meanwhile, Anthropic has already started moving away from flat-rate enterprise pricing toward per-token billing. The message is clear: the era of subsidized AI is ending.

Nobody is making money

Here's the part that should make everyone pause. The companies selling us this future are hemorrhaging cash. OpenAI's own internal projections predict a $14 billion loss in 2026. According to Deutsche Bank, the company is on track to accumulate roughly $143 billion in negative cumulative free cash flow between 2024 and 2029. As the analysts put it, "No startup in history has operated with losses on anything approaching this scale." OpenAI doesn't expect to be profitable until at least 2030. And that timeline assumes revenue somehow reaches $100 billion by 2029, up from about $3.7 billion in 2024. For context, no public company in the last 75 years has grown that fast over a five-year period. So what's the plan? Ads, apparently. OpenAI launched advertising inside ChatGPT in early 2026, starting with a $60 CPM model. Within ten weeks, that rate had eroded to as low as $25. They've already pivoted to cost-per-click pricing, competing directly with Google and Meta for ad budgets. The company projecting $14 billion in losses this year is now betting on banner ads inside a chatbot. The entire industry is running on venture capital fumes. AI companies captured 65% of all venture deal value in 2025. In Q1 2026, investors poured $300 billion into startups globally, up over 150% quarter over quarter. But as one analyst reviewing hundreds of AI startups' financials concluded, "Most AI startups will go bankrupt within 18 to 24 months. The numbers are brutal."

The universal basic income fantasy

Whenever someone points out that AI might eliminate millions of jobs, the standard response from Silicon Valley is: universal basic income. Don't worry about losing your livelihood, the government will just give everyone money. This is where the logic completely breaks down. Capitalism functions on a basic loop. People work, they produce goods and services, they earn money, they spend that money, and the cycle continues. Money circulates because labor circulates. If you remove work from the equation, you don't just lose jobs. You lose the mechanism that distributes wealth through society. UBI proposals from tech leaders tend to gloss over this. As a 2025 paper in the journal PMC argued, UBI as promoted by AI elites functions as a form of symbolic violence: it reinforces the division between those who own and control AI technologies and those who are merely consumers of its benefits. The owners of AI position themselves as benevolent providers while the recipients of UBI become passive dependents in a system they have no control over. Forbes put it more bluntly: if AI is meant to create abundance, "basic" feels like an oddly defeatist destination. UBI keeps people alive. It doesn't create social mobility, wealth, or purpose. Lock a chunk of the population into a permanent stipend while others continue to accumulate capital, and you've built a two-tier society that hardens across generations. Techno-feudalism, dressed up as progress. The Cato Institute, The Guardian, and researchers across the political spectrum have all reached similar conclusions from different angles: UBI is not an adequate answer to AI-driven displacement. It doesn't address the structural inequality that the AI economy would create.

If nobody works, nobody produces

There's an even more fundamental problem with the "AI replaces all work" narrative that rarely gets discussed honestly. Society doesn't just need work for economic reasons. It needs production. Someone has to grow food, maintain infrastructure, build housing, deliver goods, care for the sick. AI can automate many tasks, but the physical world still requires physical effort, human judgment, and coordination at massive scale. And even setting aside physical labor, work serves a psychological and social function. It gives people structure, identity, community, and purpose. The post-work utopia imagined by tech optimists assumes that people freed from labor will flourish. History suggests otherwise. Large-scale unemployment tends to produce despair, not enlightenment. The uncomfortable truth is that the "you won't need to work" narrative isn't really about you. It's about reducing labor costs for companies that need to justify their valuations to investors. When a tech CEO says AI will replace your job, they're not offering you freedom. They're telling their shareholders there's a path to profitability.

The contradiction at the center

So let's put it all together. AI companies are telling us we won't need to work anymore. But they need us to pay $20 to $200 a month to use their products. Their products are not profitable. They're burning through hundreds of billions of dollars in investor money. Their answer to declining subscription economics is advertising. Their answer to job displacement is a welfare program that even its proponents admit won't preserve dignity or mobility. And the entire thesis rests on growth projections that have no historical precedent. This isn't a vision of the future. It's a series of contradictions held together by hype, venture capital, and the hope that something will eventually work out. AI is a powerful technology. It's genuinely useful for many tasks. But the grand narrative being sold, that it will end work, that it will provide for everyone, that the economics will sort themselves out, is not supported by the evidence. The numbers don't add up. The business models don't work. And the societal safety nets being proposed are inadequate for the disruption being promised. Maybe we should stop asking "what happens when AI replaces all jobs" and start asking a more honest question: who benefits from us believing that it will?

References