AI as a 'Non-Normal' Technology: The End of the Human Bottleneck

The fundamental debate over artificial intelligence has shifted from what the tools can do to whether they can eventually replicate the full range of human capacity. This distinction determines whether AI is a standard productivity tool or a fundamentally different class of technology that renders traditional economic models obsolete.

In an essay published on '12 Grams of Carbon,' the author argues that the core disagreement in current AI discourse is whether the technology can continue to acquire capabilities until it covers the entire spectrum of human ability. The author describes this process as "smoothing out the jaggedness" of AI's current capabilities to eventually match the versatile nature of human intelligence. According to the author, the only remaining fundamental disagreement between those who view AI as a normal technology and those who do not is the belief that AI can eventually do everything a human can do, but better.

The Productivity Bottleneck

Historically, "normal technologies"—such as the steam engine or the internet—have functioned by augmenting human labor or creating entirely new categories of work. These tools multiplied the output of the human operator but still required the human as the primary driver of the process. However, a growing school of thought argues that AI breaks this pattern.

Finn Morehouse, speaking via the Ethical Machines Podcast, posits that AI is not a normal technology because it replaces the standard bottleneck to productivity—the human—rather than simply multiplying it. Similarly, Peter Wildeford has argued through the Future of Life Institute Podcast that AI's impact on forecasting and economic effects places it outside the framework of previous technological shifts. Unlike previous industrial revolutions, this technology targets the cognitive abilities of the user themselves.

Economic and Societal Implications

If AI is indeed a "non-normal" technology, the historical precedents used to predict economic adaptation may no longer apply. For decades, economists have pointed to past disruptions to argue that technological unemployment is temporary and that new jobs always emerge to replace the old. However, those precedents rely on the assumption that human cognition remains the essential bottleneck in the production chain.

If that bottleneck is removed, societal structures, labor laws, and economic systems based on human cognitive labor face an existential crisis. The shift suggests a move away from a labor-based economy toward one where the primary driver of value is no longer human effort or expertise, but the scalable capacity of the AI itself.

The Path Forward

What remains to be seen is whether AI will actually achieve the "smooth" coverage of human capacity described by the '12 Grams of Carbon' author. While current models show exponential growth in specific tasks, the transition from a specialized tool to a general replacement for human cognition is not yet a confirmed reality. Observers will need to watch whether AI continues to fill its capability gaps or if it hits a plateau that preserves the human element as a necessary component of productivity.