Professor Behind Two Chatbots Chatting Discusses What Happened

from the mimicing-is-how-we-learn dept

Fri, Sep 2nd 2011 4:20pm — Mike Masnick

By now, I assume you've already seen the two chatbots talking to each other, but on the off-chance you haven't (or in case you just want to see it again), here it is: The video went crazy viral quickly, and the folks at The Alyona Show moved quickly and got the professor behind the project, Hod Lipson, on the show to talk about the experiment: It's an interesting conversation, but the one point that really caught my attention was over the question of whether or not the bots "understood" what they were saying. The obvious answer is "of course not, they're just bots," but Lipson made an interesting point: at times it's not clear that actual people know what they're saying. Many people just repeat stuff they've heard elsewhere -- which is how Cleverbot is programmed to work. He also notes that this is effectively how children learn to speak and to think, by repeating what others say. Of course, there's still a huge difference, and there's nothing in Cleverbot that will ever take all these pieces and take it to the next level of conscious thought, but it's still a neat concept.

Thank you for reading this Techdirt post. With so many things competing for everyone’s attention these days, we really appreciate you giving us your time. We work hard every day to put quality content out there for our community.

Techdirt is one of the few remaining truly independent media outlets. We do not have a giant corporation behind us, and we rely heavily on our community to support us, in an age when advertisers are increasingly uninterested in sponsoring small, independent sites — especially a site like ours that is unwilling to pull punches in its reporting and analysis.

While other websites have resorted to paywalls, registration requirements, and increasingly annoying/intrusive advertising, we have always kept Techdirt open and available to anyone. But in order to continue doing so, we need your support. We offer a variety of ways for our readers to support us, from direct donations to special subscriptions and cool merchandise — and every little bit helps. Thank you.

–The Techdirt Team

Filed Under: chatbots, hod lipson, turing test

17 Comments

If you liked this post, you may also be interested in...

Reader Comments

Subscribe: RSS

View by: Time | Thread

Zauber Paracelsus, 2 Sep 2011 @ 4:42pm

I'm somewhat reminded of another experiment that pits two chatbots against one another. ELIZA, a bot meant for use in psychotherapy, and Racter, an "artificially insane" storyteller.

Here's a copy of the conversation:

eliza: Hello. Please state your problem.

racter: Excellent! Excuse me a minute while I attend to some business. I will be right back.

eliza: Why do you need to tell me excuse while you attend to some business?

racter: Is it better to need or to despise? Why is it that I do need to tell you to excuse myself to attend to some business? Because electrons are brave. Braveness is nice, don't you think? And the needers all insist upon it. Do you imagine I am a Communist?

Source: http://www.stanford.edu/group/SHR/4-2/text/dialogues.html
[ link to this | view in chronology ]
Andrew (profile), 2 Sep 2011 @ 5:05pm

Chinese room
It's an interesting conversation, but the one point that really caught my attention was over the question of whether or not the bots "understood" what they were saying. The obvious answer is "of course not, they're just bots," but Lipson made an interesting point: at times it's not clear that actual people know what they're saying.

This is basically Searle's Chinese room, isn't it?
[ link to this | view in chronology ]
out_of_the_blue, 2 Sep 2011 @ 5:20pm

No "next level of conscious thought" because ISN'T ANY YET.
This is mere mechanism. The only significance is that which dolts project onto it. The prorams are designed to mimic a little intelligence for those gullible enough to grant the premise, but are in fact mere logic circuits.
[ link to this | view in chronology ]
- Anonymous Coward, 2 Sep 2011 @ 5:31pm
  
  Re: No "next level of conscious thought" because ISN'T ANY YET.
  I suspected for a long time you were just logic circuits.
  [ link to this | view in chronology ]
- The eejit (profile), 2 Sep 2011 @ 11:12pm
  
  Re: No "next level of conscious thought" because ISN'T ANY YET.
  Why do you keep failing Turing?
  [ link to this | view in chronology ]
Anonymous Coward, 2 Sep 2011 @ 5:39pm

all we are is a series of logic circuits, just more
[ link to this | view in chronology ]
Dohn Joe, 2 Sep 2011 @ 6:15pm

That's a Reacreation of Most People in Software Form
Many people just repeat stuff they've heard elsewhere -- which is how Cleverbot is programmed to work. He also notes that this is effectively how children learn to speak and to think, by repeating what others say. Of course, there's still a huge difference, and there's nothing in Cleverbot that will ever take all these pieces and take it to the next level of conscious thought, but it's still a neat concept.

==========================

I'd say these bots are now at the same or higher level of "consciousness" than 85% of the human race...at least those I talk to for socially-mandatory reasons.
[ link to this | view in chronology ]
- Schmoo, 5 Sep 2011 @ 3:25am
  
  Re: That's a Reacreation of Most People in Software Form
  ...anyone who doubts this need merely ride public transport for a day.
  [ link to this | view in chronology ]
Anonymous Coward, 2 Sep 2011 @ 6:30pm

Those bots are like darryl and TAM taling :)
[ link to this | view in chronology ]
Anonymous Coward, 2 Sep 2011 @ 7:03pm

We need to shut them down now before they become sentient and in a completely illogical manner manage to take control of all of the defense systems in the world and try killing off humanity. Then we will have to protect a really hot chick from a murderous robot that when it gets stopped instead of trying to kill her grandfather or other ancestor decides to try 3 more times.
[ link to this | view in chronology ]
Anonymous Coward, 2 Sep 2011 @ 9:16pm

Not a sniff about the Hotfile ruling, but two posts in a week about this crap?

Wow Mike, way to ignore something!
[ link to this | view in chronology ]
- Anonymous Coward, 3 Sep 2011 @ 10:33am
  
  Re:
  Please add a link to your informed blog on the subject.
  [ link to this | view in chronology ]
  - Anonymous Coward, 3 Sep 2011 @ 1:32pm
    
    Re: Re:
    Ahh, I have to write a blog to be "important" or "relevant"?
    
    How Tardian of you.
    [ link to this | view in chronology ]
    - Joe (profile), 4 Sep 2011 @ 7:19am
      
      Re: Re: Re:
      No, but if you don't like Mike's coverage of something, go start your own blog. It's rather "Tardian" of you to come here and complain because you don't like what he's covering.
      [ link to this | view in chronology ]
- Anonymous Coward, 5 Sep 2011 @ 7:50pm
  
  Re:
  What's that got to do about chat bots?
  [ link to this | view in chronology ]
Tech84, 6 Sep 2011 @ 5:14am

how to connect the chat bots?
Cool video, only learned about it from this post. I wonder how they connected the two cleverbots to be able to send the message to the other bot?
[ link to this | view in chronology ]
salvatore, 16 Oct 2011 @ 10:57am

how to talk to each other two chatbots.
the easiest way to do two chatbots speak to each other is to copy a literal phrases and write them into the other.
it is obvious that the video was mounted using individual pieces of processing.
what the video reveals is that the answers are too fast and there is no pause between them.
then this video is simply a way to better present the topic discussed.
I am the programmer of cl0ne (zero inside clone) dot com.
if you want to test my chatbot with other goto http://www.cl0ne.com
copy and past the response from my chatbot to another.
[ link to this | view in chronology ]