HAL and other AIs, such as Ash or David in the Alien franchise, are spooky because they seem to have some unknown intent to them. However, this is not the case for an AI — this is a case for any team member. If we’re unable to fully understand the motives of those around or believe they have some motives counter to our own, they become scary.
Sometimes, this can be completely undeserved.
The Roomba I was roommates with started to get stuck on particular clothing items. I started to get suspicious of its motives. Why was it in our room to begin with? Why was it not maneuvering around the clothes on the floor? What was it trying to tell me?
What can be scarier with AI assistants is that their reasoning is completely a black box. What we really need is for these AI assistants to explain to us why they reach a decision. Kudos to companies like sentient.ai that have a mission to remove the black box from AI decisions.
It also helps if the AI’s voice isn’t monotone… oh, and they aren’t trying to kill us.