Wednesday, July 29, 2015

Don't ban useful robots from helping soldiers.

A group of concerned scientists, researchers and academics, including Elon Musk and Stephen Hawking, have warned that a military artificial intelligence arms race could soon develop if preventative measures are not taken.

With all due respect, these people don't work in the field so their opinion isn't as wizened nor as educated about the state of the art in robotics and artificial intelligence as they presume to be.

We are a LONG ways away from autonomous killer robots.

Autonomy is not solved.  Artificial Intelligence is not solved.  That's research speak for we don't know how to make them work. We don't know how to do either with mastery.

Here is the analogy I use.  In aviation, all flying objects work on the Bernoulli principle.  All wings develop lift from the Bernoulli effect.  All propellers develop thrust from the same effect. It doesn't matter about Wright brothers versus stealth fighters, they all work on the SAME core principle.  That is the key piece needed to make it all work.

In AI and robotics, we have not discovered the equivalent of the Bernoulli effect.  If we don't have that missing key piece, it doesn't matter how fast we make a computer, the result is still a stupid robot.

Don't let people take away useful robots from soldiers, ones that have no plans to be autonomous, by lying to you about a phantom threat.

And if you want to know how I know, go look at my Linkedin page.