## Claim AI coding tools optimize for satisfying the immediate prompt, which systematically trades away the reuse, legacy care, and honest error handling that keep a codebase maintainable. <!-- atomicity: 1 idea ✓ --> ## Scope & conditions Holds for agentic/LLM-assisted commits at scale, per GitClear's dataset and Google's independent DORA corroboration. Breaks when teams enforce reuse-first review and cause-based error handling that actively counteract the default behavior. ## Evidence Three independent GitClear metrics moved together — duplication +81%, error-masking/obfuscation +47%, legacy refactoring −74% since 2023 — and Google's DORA 2024 data corroborates the direction: every 25% more AI usage tracks with 7.2% more instability. > [!quote] > AI strongly prefers to write code that won't be labeled as a defect ## Links - [[3. Resources/MOCs/AI & the Future of Knowledge Work MOC.md]] — because this is a direct empirical counterweight to AI-coding-velocity claims ## Tests/falsifiers - This claim would be undermined by a clear counter-case where the described dynamic holds even without the scope condition above, or fails to hold even when the scope condition is met. ## Maintenance - Last tended: 2026-08-16 - Next review: 2026-11-16 ## Source [[3. Resources/Readwise/Articles/Code maintainability plummets in the AI coding era]]