DW:
Now we've just seen in the report, robots are starting to learn how to see, how to recognize, they're already better than us in other fields. When will they outperform us in terms of mind and brain?
Michael Pauen:
I think what they are likely to do or what they are likely to be able to do is to outperform us in narrowly circumscribed fields like say calculating. That's what they're already able to do. I think it's very difficult to predict and I don't think this will happen in the near future, that they will be able to outperform us with regard to social abilities. These are much more complex and much more difficult to formulate.
But in fact that there is actually a chance to talk to robotic machines, to talk to bots, clever bots and so on. So with artificial intelligence you can actually talk and this is kind of communication.
Right it's kind of a communication but that's only part of the communication; what's also part of communication is looking at each other, predicting what you're going to do, having an idea about what you are thinking about. That's what robots aren't able to do - at least so far, and I don't think that they'll be able to do this in the near future.
I can't really imagine that robots will have the capability of eyes staring at me so how will I decide whether that robot is really looking at me, with consciousness or whether it's just pretending to look at me?
Well that would require that robots are conscious and that you have reasons to believe that they are and I think we are very very far from that. I mean one reason is that we don't exactly know what it is for something to have consciousness, we have conscious experiences but we don't have clear criteria to describe consciousness to something else.
And would it actually make a difference, whether the robot just says I'm conscious or whether it really has consciousness?
Well of course, I mean you can program a robot today, I can program a stupid computer, my stupid laptop, to say I'm conscious but that's ridiculous because I don't believe that it is.
Are you scared that robots could one day take over the power of the planet?
No I'm not. People of future generations would be absolutely stupid to program or produce or make robots that have the desire to take over power and I don't think they will be.
But maybe they'll start to get autonomous, to repeat themselves and so on and then take over...
But wait they can be autonomous but these are different questions, they might be autonomous, robots are already autonomous in certain senses today but we will have to confine the degree of autonomy or the field where they can act autonomously and I think future generations will do this.
What about robots taking over our jobs? They've taken over many jobs in factories. Can you imagine a robot taking over your job, a philosobot?
I don't think that this is impossible in principle but I think it's very unlikely, it would require many resources and I don't think that future generations will spend their resources on producing artificial philosophers. Taking our jobs? I mean of course they do take jobs, that's why we manufacture machines, they're intended to do that but the manufacturing or designing of machines requires other jobs
Thanks a lot for the talk, Michael Pauen.
(Interview: Ingolf Baur)