Engineers Draft 'No AI Slop' Manifesto
The software development community has released a manifesto addressing the proliferation of low-quality AI-generated content in code, documentation, and communications. Drafted by speakers and attendees of the AIE Code Summit, the document establishes quality standards for AI usage in professional development work, drawing inspiration from how the 2001 Agile Manifesto reshaped development methodology.
The manifesto identifies “slop”—poorly vetted AI output shipped without human accountability—as a growing problem. Core values emphasize outcomes over output, understanding over speed, and developers using their own thinking rather than accepting a model’s first draft. A critical tension underlies the problem: AI tools have made content production nearly free while review costs remain constant, creating pressure to ship unreviewed work at massive scale. The manifesto notes that slop compounds over time, spreading through codebases and documentation as subsequent AI tools ingest and repeat the low-quality patterns.
Signatories commit to owning the quality and consequences of their work, reviewing AI outputs before release, and improving systems rather than repeatedly patching problems. The approach rejects pure metrics-driven measures like lines of code or merged PRs in favor of whether shipped work actually helps users. The authors explicitly acknowledge that these standards are still evolving with no single correct answer, and they invite community feedback on how to balance AI productivity with quality.
Read the full article
Continue reading at Hacker News →This is an AI-generated summary. Read the original for the full story.