- This topic has 1 reply, 2 voices, and was last updated 1 year, 1 month ago by
.
-
Posts
-
At the bottom of page 209 Peikoff discusses Rand’s idea of the “immortal robot” to illustrate that such an entity could have no values since it does not face the alternative of existence or nonexistence ie it would not be alive. I’m wondering if this can be contrasted to (unreasonable?) concerns about AI. According to one definition AI “involves using computers to do things that traditionally require human intelligence”. If I understand it correctly, concerns about AI are that it could take self-generating action that would harm humans and that we would not be able to control it. Do such concerns effectively rest on the assumption that AI would not only be alive and conscious but would, similar to humans, have the capacity for volitional conceptual thought? If so, are these concerns warranted?
I think the main concerns are that AI could be an extremely powerful tool, and that evil people could put it to evil use, for instance, asking ChatGPT how to build a nuclear bomb.
I think AI is distinct from the immortal, indestructible robot in that it isn’t immortal or indestructible. If an AI loses power, it ceases to exist. So, if somehow AI developed consciousness, it would have to have values.
The indestructible robot isn’t (in my view) the most helpful example, because it is a hypothetical that defies all known laws of existence. The closest actual existent to it would be a rock or a rubber tire or some other inanimate thing.
- You must be logged in to reply to this topic.