Artificial intelligence is no longer a monolithic force. Instead, it's splintering society into three sharply divided groups with fundamentally different relationships to the technology, creating what amounts to a new kind of economic and cultural fracture.
The first group consists of power users who have moved well beyond casual experimentation. These are people running AI agents continuously, swapping strategies for automating work and decision-making, and racing to use up their monthly token allotments. Former OpenAI and Tesla AI researcher Andrej Karpathy recently revealed he spends 16 hours daily commanding AI agent swarms. He noted that many people's entire understanding of the technology comes from a single free trial of ChatGPT, a snapshot that fails to capture what advanced systems can actually do.
A second, much larger group remains skeptical. These doubters see AI through the lens of glitchy chatbots and viral failures. They haven't engaged with the technology's full capabilities and maintain a wait-and-see attitude, viewing it primarily as a curiosity rather than a tool reshaping their work and lives.
The third group is smaller but growing louder: deliberate resisters who understand exactly what AI can do and actively oppose its development. This camp views the trajectory of the technology as dangerous and wants nothing to do with it.
The gap between advanced users and everyone else is widening measurably. Anthropic's March economic impact report found that experienced users attempt more complex tasks and succeed more frequently, creating a self-reinforcing cycle of productivity gains. Box CEO Aaron Levie called it
Comments