deslop-detectlisted

# AI Detection Prevention AI content detectors measure statistical properties of text — not meaning. They look for patterns that distinguish machine-generated sequences from human-written ones. Understanding what they measure tells you exactly where to intervene. ## When to Use ✅ Use for: - Content that was partially AI-assisted and needs to pass detection for legitimate publishing - Marketing copy that AI detection is flagging incorrectly (false positive remediation) - Testing whether desloped content will clear detection thresholds before publication - Understanding why a piece is flagging and how to fix it - Ensuring compliance with platforms or clients that require human-written content ## When NOT to Use ❌ NOT for: - General copy quality improvement without a detection concern — use `deslop-copy` - UI microcopy — use `deslop-ui` - Academic fraud, plagiarism detection evasion, or submitting AI work as human in academic contexts - Deceiving audiences about the fundamental origin of content in ways that harm trust - Code cleanup — use `ai-slop-cleaner` ## How AI Detectors Work The three metrics that most detectors measure, and what they actually check: ### 1. Perplexity How predictable each token is given its context. Language models predict the most statistically likely next token. When a model writes, it produces low-perplexity text — everything that follows is what you'd expect. Human writers make less predictable choices: unusual word order, unexpected transitio