The problem is not AI code, but not knowing about system architecture or intent
Source
ssp.sh
Author
zazuke
Date
Context
When a team delegates implementation to AI and nobody retains why the system was designed the way it is, maintenance becomes hard. That is the argument of this essay on architectural knowledge: the problem it names is not the quality of the generated code but the loss of the intent behind design choices, which later changes depend on. The essay is an experience-based argument supported by anecdotes and quoted comments from engineers, not a measured finding that AI necessarily reduces engineering knowledge.
Terms in this piece · Glossary
AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
