Hacker News Reader

HNHacker News

Top New Best Ask Show Jobs

Comment by WhitneyLand | Hacker News Reader

Parent Full thread

WhitneyLand·

Solved zero shot.

Next?

https://chat.openai.com/share/91392131-90ff-45ab-8ea4-963f73...

View on HN

It failed at the first step. This is like the worst timeline where people just cannot think for themselves and see that the AI produced an answer and so it must be true.

you’re reading way too much into my post.

It’s lots of words all run together for the purpose of being a logic puzzle and obviously I made a parsing mistake in my brain.

I’m not trying to assume AI is right, I’m trying to put a factual stake in the ground, one way or the other so we have more data points rather than speculation.

I dunno. Don't you think this could happen with other replies from ChatGPT? I think this is the "it" about this tech - it really, really does trick us some times. It's really good at tricking us, and it seems like it is getting better!

First, what custom prompt did you use? "This conversation may reflect the link creator’s personalized data, which isn’t shared and can meaningfully change how the model responds."

Second, it isn't even right:

Third Trip to the Woods: The person takes the balloon to the woods. Now, the person, the vacuum cleaner, and the balloon are safely in the woods.

In the very first step it leaves the balloon alone with the vacuum which is illegal.

"First Trip to the Woods: The person takes the magical creature to the woods first."

True!

I'm confused, in your example it immediately got it wrong by leaving the vacuum cleaner and balloon together, and then does it again in step 6.

Hilarious. People are so confident in ChatGPT that as soon as they see a plausible-sounding response it must be correct. In a discussion about proving ChatGPT has intelligence... maybe we need to prove humans have intelligence first.

Reply on news.ycombinator.com