Yesterday: I Predict Killer Robots Will Be Justified By Companionship

New York Times, today.

Despite his concerns, Dr. Horvitz said he was hopeful that artificial intelligence research would benefit humans, and perhaps even compensate for human failings. He recently demonstrated a voice-based system that he designed to ask patients about their symptoms and to respond with empathy. When a mother said her child was having diarrhea, the face on the screen said, “Oh no, sorry to hear that.”

Great. Just great. Robots' great advantage: wealth creation and life extension. Oh yay, some version of long-lived super-rich people destroying irrelevant, non-wealth producing inferiors with flesh-eating robots.

And now, the serious point. You have to assume with any technology that it will disseminate to criminals and unethical political actors without severe controls. Worse, it can prove very hard for otherwise decent minded people to resist a temporary rapid benefit, and lose something much bigger. An example is nuclear weapons; Edward Teller's increasingly insane schemes, like the serious attempt to use H-Bombs to dig out ports, came along with plans for an atomic doomsday plane that spit out h bombs and lethal radiation indefinitely. They had to be fought.

The world is heavily stressed by nuclear techonology control as it is, and it still isn't working well, but this is an example of what I have come to think should be done. As the economic and social power made possible by advanced technology rises, the day of very strong, international treaty-level controls on AI, bio-tech, and robotics is coming.

And I might add that I am still waiting for strong evidence that we will survive the invention of the electric lightbulb and the internal combustion engine.