CONCEPT
AI-Generated Code as Ultimate Abstraction
The class of software produced when a developer describes intent in natural language and a language model returns working implementation across the full technology stack — the most powerful abstraction ever built, and the one whose structural leak profile Spolsky's law predicts with uncomfortable precision.
AI-generated code names the category of software whose implementation is produced by a large language model in response to natural-language specification. Introduced at scale with Claude Code and its competitors in 2024–2025, the practice collapses the traditional software development workflow — specification, design, implementation, review — into a conversational loop between human intent and machine output. The result is the most radical productivity multiplier in the history of software, and also the most radical expansion of the gap the Law of Leaky Abstractions measures. Where previous abstractions hid one layer from the layer above, AI-generated code hides every layer simultaneously, producing the conditions for leaks whose severity is proportional to the gap between the developer's understanding and the system's complexity.
In The You On AI Field Guide
The mechanism is straightforward: the developer describes what the software should do in English — her own language, with all its ambiguity
Keep reading with YOU ON AI
Unlock the full book, 10,000+ field-guide entries, and a 1000+ thinker library. If you have a book code, register now — it takes a minute.