Taken from the Philosopher AI website :

This is an experiment in what one might call "prompt engineering", which is a way to utilize GPT-3, a neural network trained and hosted by OpenAI.

GPT-3 is a language model. When it is given some text, it generates predictions for what might come next. It is remarkably good at adapting to different contexts, as defined by a prompt (in this case, hidden), which sets the scene for what type of text will be generated.

Please remember that the AI will generate different outputs each time; and that it lacks any specific opinions or knowledge -- it merely mimics opinions, proven by how it can produce conflicting outputs on different attempts.

by Murat Ayfer

philosopherai.com