Cognitive Singularity …


Last week I had a discussion with a colleague about the actual developments in AI, mostly around the ongoing cognitive buss. I asked her if she knows ELIZA, which she answered no to. So I explained that ELIZA   was created by Joseph Weizenbaum to demonstrate the superficiality of communication between man and machine. Eliza simulated the conversation of a psychotherapist and a patient in an initial psychiatric interview. Here is a later implementation: http://www.med-ai.com/models/eliza.htmlELIZA_example Eilza was regarded as one of the first programs capable of passing the Turing Test. The discussion went on and we touched the topic of CAPTCHA which sometimes is described as a reverse Turing test and should prove that the responder is a human and not a program. I wonder myself now, how close we are to the point of technological singularity (see some thoughts about this in my blog post here) when we already are not testing if a human being cannot identify if talking to a computer program or another human being but testing if a computer program can decide if a human being is answering and not a computer program? Seems that we are at least close to some cognitive singularity here …

 

2 comments

  1. Nurcan

    Thank you for sharing these interesting thoughts, Lutz. However, look at all the “communication accidents” happening on a daily basis – in our microcosmos and in society. How often in even such simple conversations exchanging 3-5 word-sentences on ordinary topics one may experience painful surprises: all in a sudden you realize that you said something, that your partner understood in a completely different way and in turn you hear things, you would not have thought possible to be said. I may be completely wrong, but I am not sure if we should not better (or at least: first) invest our resources in improving inter-human conversation, before trying to teach our machines a virtue that we do not properly master ourselves. [let alone that conversation with a real human being will sometimes convey conversation with a machine….]

    • Dear Nurcan, absolutely agree, we should invest in improving inter-human conversation, yet this will need to be a parallel development path, as human-machine conversation is a very important and interesting area for many technical developments. For the inter-human aspects of communication, I very much like the communication square model by Friedemann Schulz von Thun https://en.wikipedia.org/wiki/Four-sides_model to start with. Reflecting on this model can help a lot to understand why communication went wrong. A second approach is addressing the mindset. I like one publication very much by the Arbinger Institute http://arbingerinstitute.com/ It is called the”Outward Mindeset” https://goo.gl/ffiJrn which addresses why the concept of thinking outside of ourselves is critical in conflict resolution.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: