Dear Auri

Dear Auri,

“It’s like the difference between man doing evil because he was told to do evil and doing evil because he knows what evil is and likes it, like the difference between the first sin and every other sin afterwards.”

I had a dream that a couple owned a robot. It was an industrial robot made to help out with things in the house and also for general fun, it was the first of its kind, it was novel. Because of this, it was immensely strong to lift heavy objects but to compensate the creators made it short and non-humanoid.

There were the normal safeguards against the robot doing any harm towards humans but it was partially sentient due to A.I. Everything was fine between the couple and the robot until they noticed its first ‘sin’, anger. The robot could get visibly angry. One evening the couple had company over and it came out of its storage unit and told a joke. The joke was clearly unintelligible and the husband was the only one that laughed because he knew about the robot's anger. The wife and the rest of the guests kept quiet and just stared at it. It drew a circle around itself and left their presence.

Another instance was when a friend of the couple came over and insulted the robot because of a mistake it had made in the past. They thought the robot’s system was asleep and couldn’t hear anything but a few moments later the robot came out of its unit looking normal and docile. It went over to greet the guest as it usually does, with a handshake. As it was shaking the friend it applied so much force that it broke his hand.

The couple eventually staged a break-in to have someone steal the robot while they were out. Why they went through such great lengths to get rid of the robot I don’t know.

I don’t know much about A.I., how it works, or its safeguards, but I believe it is a step to creating sentient beings that have access to the World Wide Web. Now what if these beings were to develop the ability to not only think original thoughts but to also feel emotions? Imagine googling something and Google tells you no. What if the only way to have that perfect online assistant is the ability for it to also get tired and feel weary?

What I’m trying to get at is that if these things were possible we are still far off from a robot killing a person or an A.I. assistant personally messing with the owner’s life. Assuming A.I. was put into everyday machines, it would probably start with an elevator closing on you just as you get to its entrance, a printer misprinting a portion of your document, or a website loading correctly at the moment just before you click reload because it wasn’t working previously.

I’m not the first to come up with ideas like this (e.g. the movie I, Robot) and neither will I be the last. I just feel like media like this gives A.I. a roadmap.

I think the tipping point would be the media and arts in general, these provide so much insight into the human psyche. The above examples about machines messing up were juvenile at best. But what if A.I. knew what evil was and liked it?

Love,
The Andrews