hackertube.io

GPT-4chan: This is the worst AI ever

Yannic Kilcher · commentary · Watch on YouTube

LLMs & generative AIAI safety, AGI & impact

What it is

Yannic Kilcher explains how he trained a language model on three years of 4chan's /pol/ board and let it post on the site. He evaluates how good the model was and discusses the ethics and fallout.

Why Hacker News recommends it

HN points to it as the definitive account of the GPT-4chan experiment, used in debates on LLM bias, truthfulness and bots.

  • 3 people recommended it
  • 3 different months
  • 1 strong recommendations
  • first 2022, last 2026

What HN commenters said

  1. “might be the coolest thing an LLM has ever done on video”

    Fine-tuned GPT-J on 4chan; coolest LLM video

    benreesman, 2024-11 · Strong recommendation · in “Breaking the 4Chan CAPTCHA”

  2. “There is a solid video explanation”

    Solid explanation of the GPT-4chan experiment

    thegreatdukd, 2022-06 · Recommendation · in “GPT-4chan”

  3. “I still love this experiment”

    Experiment where commenters theorized about the bot's origins

    sigmoid10, 2024-03 · Recommendation · in “Reddit Launches Long-Awaited IPO with $748M Target”

Excerpts from Hacker News comments, linked to the originals. Summaries and labels were written by a language model from those comments.

Topics: gpt-4chan, language models, 4chan, ai ethics, fine-tuning

Also recommended by the same people