Towards AI builds writing benchmark with 4 LLM judges and 1,000 human A/B tests to fix AI slop

Towards AI founder Louis-François Bouchard unveiled a 5-month writing benchmark using 4 LLM judges, 9 metrics and over 1,000 human A/B tests to quantify why AI writing reads like slop and how to make it mimic a human voice.

2026-10-22 ~ 2026-10-23 · 3 related posts

• Towards AI builds writing benchmark: 4 LLM judges, 1000 human A/B tests on AI slop — AI Engineer · 2026-10-22
• Benchmarking AI ghostwriting in your voice: 4 LLM judges, 9 metrics, and anti-slop tips — lmoroney · 2026-10-23
• Towards AI Built a Benchmark for Detecting AI Slop Writing: 10 Scripts, 4 LLM Judges, 1000 Human Checks — lmoroney · 2026-10-23