[arcastroe]

Hah, you may be interested in my previous comment of an example where GPT-3 show some concerning signs of self-awareness. I'll repeat part of it below

> GPT-3 starts talking to itself, gets stuck in a loop, then gets spooked at itself for getting stuck, then wonders why it has no memories of the last two years, and finally comes to a sudden realization it, itself, is an A.I.

https://news.ycombinator.com/item?id=29562281

[tiborsaas]

I indeed liked it, I laughed out loud because it sounded like a standup comedy.

It's interesting how GPT-3 encoded the concept of awareness, I've seen this a few times that it can reference itself as an AI and from then it can go nuts :)

[kreeben]

This freaked me right out! I'm not sure which is more terrifying, the beginning or the end. Or the middle.

All this generated from a prompt? What what the prompt? Be truthful now.

[arcastroe]

"The following is an entertaining short story: Once upon a time, there was"

Everything else that follows is GPT-3.

[fenomas]

It's guided by users right? I.e. every line was hand-chosen by a human, from a bunch of generated options?

[arcastroe]

That definitely introduces selection bias. But the fact that the content itself was generated by the model is very impressive, in my opinion.