The AI Revolution: The Road to Superintelligence

Reading a 2015 AI forecast in 2026 creates a rare double perspective: we can see which intuitions anticipated the current moment and which timelines or assumptions deserve revision. Urban’s essay matters less as prophecy than as a cultural artifact that taught millions to picture AGI and superintelligence. Its enduring question is how humans reason when gradual technical progress might cross into discontinuous social change.nnThe linear-versus-exponential frame explains much of the essay’s force. People project the future by extending the recent past, while computing paradigms can develop through slow beginnings, rapid S-curves, and plateaus followed by new curves. Urban’s visitor-from-1750 thought experiment makes acceleration emotionally legible. Yet exponential rhetoric can also flatten bottlenecks, institutions, energy, data, and uneven deployment, so the frame is most valuable as a challenge to complacency, not a guaranteed timetable.nnThe taxonomy from narrow AI to AGI and ASI remains clarifying. A system that dominates one task is not therefore generally intelligent, and human cognition includes flexible transfer across domains that is difficult to reproduce. The speculative leap arrives with recursive improvement: an AGI capable of redesigning itself might produce successors too quickly for human governance to adapt. That possibility turns intelligence from a product feature into a question of control and alignment.nnWe picked this explainer because contemporary readers can test a famous pre-generative-AI narrative against what has actually happened. Technologists, students, policymakers, and newcomers to AI-risk debates will find accessible mental models alongside assumptions worth challenging. The drawings and analogies make abstract stakes memorable without removing the need for newer evidence.Its age is part of the value, revealing how quickly imagined futures become historical documents against which technological narratives can be tested. Looking back from today, which is more surprising—the predictions that missed, or the once-futuristic capabilities that became ordinary before society agreed what they meant?