Haves, have-nots and know-nots: Inside AI's new class divide

According to Axios, a striking class divide is shaping how Americans experience artificial intelligence. For frontier power users—developers, researchers, and tech elites—new models like OpenAI's Sol and Anthropic's Fable represent a revolution: autonomous coding agents solving complex problems with minimal intervention. But for most Americans, AI remains an evolution: a smarter search bar, a faster inbox. Nearly half of U.S. adults now use AI chatbots, yet mostly for basic search—the same job Google has done for decades. The population running agentic coding tools is a fraction of that fraction. And trust is slipping: sixty-three percent of Americans say AI is advancing too quickly, while just sixteen percent expect it to benefit society over the next twenty years. Even among elites, there's a hierarchy: free tiers, paid subscribers, power users, preview access, and an insider class testing unreleased capabilities. A century ago, electricity exposed a similar divide between urban and rural America, until the New Deal's Rural Electrification Administration bridged it. AI's divide may prove harder to close, as frontier access remains scarce and expensive.

Listen to this story

Hear this and more stories in a personalized audio briefing.

Open The Chonkerton