
A.I.’s Original Sin
The Daily
Limitless Ambition: Never Satisfied
In late 2021, OpenAI faced a challenge as they exhausted almost all the English language text available on the internet to train their chatbot system, GPT. Despite having consumed billions of words from media articles, news, Reddit threads, and digital books, they desired even more data to enhance their system further.
00:00
Transcript
Play full episode
Remember Everything You Learn from Podcasts
Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.