What metrics beat perplexity for detecting AI-generated content?

Zipf's Law conformity, repeated sentence starters, punctuation entropy, sentence skew, hapax ratio, and paragraph uniformity — clustered multiplicatively per sentence.

How accurate is this browser-based AI content detector ?

Benchmarks show it crushes where perplexity flops, but no 100% — modern LLMs fake human traits well; clusters catch patterns others miss.

Can I use this AI detector for free without sign-up?

Yep, fully browser-based — no uploads, no subs, no APIs. Paste and scan instantly.

🤖 AI Dev Tools

Why Perplexity and Burstiness Failed — And What Actually Catches AI Slop

AI detectors promising the moon with perplexity and burstiness? They're yesterday's news. A new browser-based sniffer nails modern LLMs where others flop.

theAIcatchup Apr 10, 2026 3 min read

Browser screenshot of AI content detector analyzing text with score readout

⚡ Key Takeaways

Perplexity and burstiness fail on 2025 LLMs; they're too human-like now. 𝕏
Zipf's Law deviation and signal clusters are the real AI detectors. 𝕏
Browser tool spots GPT/Claude fakes via 10 stats + 18 sentence signals — free, no cloud. 𝕏

Published by

theAIcatchup

Ship faster. Build smarter.

#AI content detector #AI detection #Zipf's law #burstiness #content authenticity #perplexity

Worth sharing?

Get the best Developer Tools stories of the week in your inbox — no noise, no spam.

Originally reported by dev.to

⚡ Key Takeaways

The 60-Second TL;DR

theAIcatchup

Share this article

Worth sharing?

Related Stories

AI Text Detectors Exposed: Perplexity Tricks and Burstiness Busts That Don't Hold Up

Cloudflare Scans 3.5 Billion Scripts Daily — Now Free, But Is It Foolproof?

One Forgotten Line: How Anthropic Handed Rivals Their $340 Billion AI Crown Jewels

7 Prompt Engineering Tricks That Haven't Died Yet in 2026

Stay in the loop