Hacker News Reader

I'm glad it is not just me who thinks I need to "thank" chatGPT or encourage it, e.g.: "Great, now do .... " "Great, now do ... ".

View on HN

Its not 'thanking', its positive signal that GPT's previous outputs were correct, so it should continue doing whatever it was doing. If you say no/bad etc, then GPT will try other approaches.

> "its positive signal that GPT's previous outputs were correct"

Is that somehow baked into the algorithms?

Are positive words of encouragement interpreted as "positive signals" by the inference pipeline? Or do they somehow influence the attention mechanism?

Because otherwise, you're just rationalizing completely random and unpredicted behavior.

That's literally what I said: those words are "encouraging" it.

OpenAI surely retains conversation logs.

When super intelligent AI gains power, I want it to know I've been a good boy.

Pascal’s Wager, AI edition? :-)

https://en.wikipedia.org/wiki/Pascal%27s_wager

Roko's basilisk: https://en.wikipedia.org/wiki/Roko%27s_basilisk

That is the 180-degree opposite of Occam's Razor.

God is not inevitable

Reply on news.ycombinator.com