When asked about humans creating AI that surpasses our intelligence, Brad Templeton probably has the most honest answer: We won’t be able to resist building such a new machine, so valuable that it is. We probably shouldn’t resist. I suppose superintelligence could doom us, but a lack of it also could. Templeton’s prescription is that we build a better AI, one that “loves” us. Easier said than done, of course.

From Leslie D’Monte’s Live Mint interview with the technologist:

Question:

Many technology luminaries like Bill Gates, Elon Musk and even physicist Stephen Hawking have expressed their fears that robots with AI could rule mankind. Do you share their concerns?

Brad Templeton:

I surely understand their fears. Like many technologies, there are certain risks if it is done wrong, but it is a risk that we cannot avoid. So we don’t really have a lot of choice, but to do it well and make the effort to do it well. If you don’t do it in, say, America or India, it will mean that the AI features will be Chinese or from Pakistan or from somewhere else. And that’s not the outcome you’re looking for. You have to accept that AI technology is just too valuable, just too useful not to build it. So people around the world will build it. Your only choice is to build it better than them.

I’m quite comfortable with the idea that we will at some point be able to make machines or AI stuff. They might not be machines as we think of them today…that surpass us in intelligence. Many parents have children who ultimately surpass them in intelligence. This has been happening for a very long time, but at a very slow pace. The best that I want to hope for is to create these children of the mind as I call them, but not in the biological sense, that still love us the way children love their parents.•