Robot Wars

In another post, Noel Sharkey and I have been debating the control or otherwise of robots designed/presumed to care for children. I stand by my mistrust of legislation and top-down guidance on topics like this, preferring education and individual accountability. But when it comes to military robots, Noel and I are on pretty much the same wavelength, so I wanted to make a separate post about that to alert people to the topic and the issues he raises.

When it comes to the use of stupid but autonomous robots for military applications, I agree with Noel in large part. So I suggest you read his IEEE Intelligent Systems magazine article here.

I don’t pretend to know what we can do about this. It’s one of those “if we don’t do it, they will” situations, and those are very dangerous feedback loops that cause people to do things they know to be crazy.

Ironically, if we ever get genuinely intelligent robots, with intelligence on a par with humans, then I’m convinced they’ll qualify as moral beings themselves and the problems will get easier. The currently looming quandaries only apply to stupid automatic systems. We have had stupid automatic weapons for a long time – mines are an obvious example. We don’t even expect these to discriminate between military and innocent targets, but once their behavior becomes more conditional and they are expected to make decisions and believed to be right, a lot of dangers arise.

I do hope people can keep these ideas separate from the general fear of robots, however, since the latter is misguided. We’re not talking robotic warriors of the Hollywood variety here. In fact it’s their complete stupidity that’s the problem.


About stevegrand
I'm an independent AI and artificial life researcher, interested in oodles and oodles of things but especially the brain. And chocolate. I like chocolate too.

7 Responses to Robot Wars

  1. Herni says:

    I think that as long as we don’t get to create machines with authentic intelligence, this is just simply off topic. Human-made devices will always be in the same category as guns or dolls unless we got to the point where actual intelligence is developed. I think it’s.. hmm.. ‘healthy’ to see that there are people eager to discuss about moral issues regarding AI, but it’s too soon to do it regarding the scarce achievements we’ve done. I agree that present legislation is enough to judge wrong behaviors; no one should trust in a machine as in another human being. If someone does that, that person is the one to blame for any undesired consequences.
    This debate should be resumed once we hit the milestone of creating a conscious machine. Only then the actual legislation would be obsolete. As of today, military robots are just complex guns and robotic child care devices are complex dolls made to keep babies entertained.

    -Sorry for my English-


    • stevegrand says:

      Thanks Herni! This was bugging me all night because I seemed to be in danger of contradicting myself, but you’ve put me straight. This is an issue about Weak AI, and I don’t do Weak AI.

      These potential military problems are a consequence of Weak AI research – “enabling computers to do what humans normally use intelligence to do”. They involve task-oriented decision making systems, which is NOT the same thing as true intelligence. I work in “artificial intelligence through artificial life”, as Larry Yaeger describes it, which is a very different thing. Nobody in their right mind, even the military, would use my kind of AI to make decisions involving guns and bombs. It’s precisely equivalent to giving a gun to an animal (a dog, say, which would be madness; or, in the very distant future, a human, which is something we already do).

      The problems arise because Weak AI allows us to mimic certain aspects of intelligence, up to a point. There are genuine dangers in this and Noel is right to be concerned. But people mustn’t extrapolate too far, because this Good Old-fashioned approach to AI (and even most of the insect-inspired New AI that is currently being heavily funded by the military) will inevitably hit a brick wall.

      Unfortunately, the general public (and even many AI researchers) lump these two radically different approaches to AI together. Which means, as a member of the set “AI researchers”, I have an obligation to take part in the debate. But perhaps my most useful contribution is to continue the fight that has already cost me two books and a hundred public lectures, to explain the differences between automated decision making and living intelligence.

  2. Herni says:

    I would have sworn the Sim-biosis site would be working before the Grandoids one was up.
    Congratulations, Steve. Beautiful design, really, it’s a very appealing site.

    I guess this new project means you are now broadening your field of work: “Grace will not have to demonstrate general intelligence.” Your goals are now different and probably the means are to be different too (guess this also means you are now entitled to discuss about weak AI :P). I’m sure, though, that you will know how to use this to learn things that will be of use with respect to artificial life and AI. Not to mention this will hopefully help fund your research in that matter.

    So best of luck to you with this new endeavour.


  3. stevegrand says:

    I’ll pass the congratulations on to Sara, Herni, since she’s the web developer. Thanks!

    Yeah, I guess I am selling out a bit. Sorry! I’m using an A-life approach and neurological inspiration as much as possible, but inevitably we’ll have to cheat heavily when it comes to speech, etc. I’m treating the project largely as an exercise in cybernetic art, rather than science or technology, although the robots will make useful research tools in their “spare time”. But the main objective is to earn some actual money for once. Never mind research funding – some cash to pay the mortgage would be nice!

    Thank you so much for joining our fledgling community already!!!

    The Sim-biosis site may be up (to some extent) in a few weeks. Sara will soon be ready to work on it but I’m so busy with the robot that I just don’t have the brainpower left to think about the game and the content for a website right now. I’m not sure if I got my priorities right but something had to give.

    Thanks for the kind wishes!

  4. Mark Jones says:

    Dear Steve,

    Please excuse me for using this forum to ask you a completely unrelated question.

    Richard Dawkins intriguingly quote you in his recent book, The God Delusion, “not a single atom that is in you body today was there when that event took place…” meaning, that every atom in the human body is replaced within our lifetime. I have read this assertion many times, but it’s source is never attributed. It is a facinating if true but I don’t want to quote it unless I’m sure of its scientic origins. Who can be cited?

    Thanking you in anticipation.

    • stevegrand says:

      Hi Mark,

      Thank you for the question! I’ve lost count of the number of people who’ve asked me this – they tend to want to believe it but are unsure of its veracity. Even Richard double-checked with me because people had been asking him about it. So I reckon it’s worth a whole post to open the discussion up to anyone interested. I was wondering what to blog about today…

  5. Pius Agius says:

    Dear Steve,

    I just replaced a faulty printhead on my printer and I feel that I can fix anything. Well actually that is a bit much. It is amazing how long I took the other day to get my printer to print. I tried everything and still no luck. A call to the servicing department and the kind technical helper nailed it quickly and shipped me a new part.
    Well what has this got to do with robots at war? We, I mean humans, rely on our tools, low-tech like screwdrivers and high-tech like phones and computers. Now we are at the development stage of a wonderous new technology. Robotics, I feel will actually help mankind with the problems we face and seem never to completely go away.
    Saying that I am very, very upset when I see robots being used to destroy or injure those they are supposed to be aiding. Granted, at the present stage robots are still being developed. They are like seeds we plant. What will be the end result if the only substantial money is being provided by the military?
    I hope in building such machines we become wiser. Time will tell and I just had to get my two cents in. I only hope it makes some sense.

Leave a Reply

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

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

Google photo

You are commenting using your Google 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 )

Connecting to %s

%d bloggers like this: