A Simulation in C++ of Joseph Weizenbaum's 1966 Eliza

#1

Post by m1guelpf » Sat, May 17, 2025, 4:58 PM UTC

Re: A Simulation in C++ of Joseph Weizenbaum's 1966 Eliza

#2

Post by johnea » Sat, May 17, 2025, 8:15 PM UTC

And how do you _feel_ about LLMs not constituting any kind of "intelligence"?

I'm sorry you feel that I'm over anthropomorphized...

#3

Post by tantalor » Sat, May 17, 2025, 8:48 PM UTC

> purpose was not to advance our understanding of intelligence, but to... demonstrate how easily people can be fooled

That actually says a lot about "intelligence", doesn't it?

#4

Post by Animats » Sat, May 17, 2025, 9:17 PM UTC

At the time, there was said to be a seduction script for the Eliza engine. Has it ever been found?

#5

Post by wpollock » Sat, May 17, 2025, 9:48 PM UTC

Animats wrote: post #4

At the time, there was said to be a seduction script for the Eliza engine. Has it ever been found?

Yes it was, handsome!

#6

Post by throwaway81523 » Sat, May 17, 2025, 11:22 PM UTC

You can run a Lisp version with M-x doctor in Emacs.

#7

Post by quuxplusone » Sun, May 18, 2025, 2:21 AM UTC

This footnote on the famous Parry/Doctor conversation is interesting and deeply researched: https://github.com/anthay/ELIZA/blob/master/doc/Trying_to_re...