If I am the human end of the Turing Test, how do I make it harder for the computer? [closed]

Question

Closed. This question needs to be more focused. It is not currently accepting answers.

---

Want to improve this question? Update the question so it focuses on one problem only by editing this post.

Closed 8 years ago.

Improve this question

Suppose I've been recruited to a university study testing a new AI, and we're going to do a traditional Turing test. I will sit at a computer and interact via text-based chat with something that is either a computer or another volunteer, and then have to guess which it is.

Suppose I've been given a particularly nice incentive to get it right. How can I improve my odds? What sort of things should I talk about?

• Should I try to teach the chatter a simple new concept I just made up, and then quiz them about it?
• Should I try to make jokes and innuendo and see if they "get it"?
• Should I ask philosophical questions?
• Should I try practical questions that humans tend to "irrationally" answer in a particular way?

Answer 1

Realize that the state of the art in chatbots is still abysmal. The biggest mistake you can make is to try to do something bizarre, in which case both a human and an AI might reasonably respond in a bizarre manner. Instead, just engage in a normal conversation and resist any significant deviations proposed by your correspondent. That should easily let you identify anything operating at the current published state of the art.

If you encounter something well beyond state of the art, I think the next threshold would be a directed inquiry into a aesthetic decision: "What's your favorite [movie|book|artist|meal]? Why? So would you say you never like...? Are there aspects of that you dislike...?" etc.

Answer 2

Try some l33t speak, that might do it. Then again I dont even understand some of it at times. Maybe I'm a computer. Maybe you're a computer

Answer 3

Since computer programs might be very logical and high performing, but have no experience of being a human being, I would suggest questions of type trivial everyday events that wouldn't get adequate answers on Google, likely. As questions about cleaning, small details on outfits, boring and irritating situations ...

And it might be a good idea to communicate in a spoken careless way with lots of words and sentences with clear and controversial statements of the above type only here and there.