
4chan Gamers Discovered AI Chain-of-Thought Reasoning Before Google
New research reveals 4chan users accidentally discovered chain-of-thought prompting in 2022, over a year before Google researchers claimed to find it first. Learn how grassroots AI experimentation beat formal research.
What Is Chain-of-Thought Reasoning? — AI's Biggest Breakthrough
Chain-of-thought (CoT) prompting is the technique of asking AI to solve problems step-by-step rather than jumping to answers. It's widely considered one of the most important discoveries in AI prompting, dramatically improving model performance on complex tasks.
Google researchers were credited with discovering this in 2022. But new research from The Atlantic reveals that 4chan users found it first — by accident, while playing a game.
How Did 4chan Users Find It First? — Accidental Genius
In 2022, 4chan users were playing AI Dungeon, a text-based game powered by GPT-3. They started asking AI characters to solve math problems and noticed something interesting: when they asked the AI to work through problems step-by-step, accuracy improved dramatically.
They didn't write a research paper. They didn't call it "chain-of-thought reasoning." They just noticed it worked and moved on. Meanwhile, Google researchers made the same discovery independently and published it, getting all the credit.
What Does This Tell Us About AI Research? — The Innovation Gap
This discovery highlights a fundamental truth about AI: the people using AI daily often find breakthroughs before the researchers studying it formally. The gap between practice and theory in AI is enormous.
The gaming community, modders, and hobbyists have always been early adopters. The fact that one of prompting's most important techniques was discovered in a game chat room says something profound about how innovation actually happens.
FAQ
Q: Did Google copy the 4chan discovery? A: No evidence suggests Google was aware of the 4chan findings. Both appear to be independent discoveries, but the 4chan users found it first.
Q: What is AI Dungeon? A: AI Dungeon is a text-based adventure game powered by GPT-3 where players interact with AI characters in open-ended scenarios.
Q: Why didn't the 4chan users publish their findings? A: They were gamers experimenting for fun, not researchers. They observed the phenomenon informally but didn't document it in academic channels.
Stay ahead of the AI curve. Follow @AiForSuccess for daily insights.
📬 Want more AI solopreneur insights?
Subscribe to our weekly newsletter →Related Articles

AI Model API Aggregation Platforms: From Simple Proxies to Enterprise AI Hubs
AI API aggregation platforms are evolving beyond protocol translation. Discover how these platforms are becoming essential infrastructure for enterprise AI adoption.

AI Jobs Explosion: 12x Increase in AI Positions Signals Massive Talent Demand
The AI job market has exploded with a 12x increase in AI positions since 2025. Discover what's driving this talent rush and what it means for your career in artificial intelligence.

Anthropic's Claude Code Source Leak: 1900 Files, 500K Lines of Code Gone Public
In June 2026, Anthropic accidentally published nearly 1,900 source files and 500,000 lines of Claude Code's core codebase. Here's what the leak revealed and why it matters for AI developers.