If a computer is capable of handling the given task, how far would you trust it? For example, would you trust it to drive a car? Run self-replicating robots? Lead a country?
Page 1 of 1
How Far Would You Trust A Computer?
#6
Posted 05 September 2008 - 01:49 PM
Self-replication seems to me the task that would require the most trust, it's the one that all the sci-fi horror stories are based on. Whether it's the nano-machine 'gray goo' issue, or the usual making lots of robots to overthrow humanity. Then again if we develop sentient or near-sentient computers and still treat them as the utilities we do currently, we probably deserve whatever we get.
If the technology is reasonably sophisticated, I would trust them to take over many of the mundane tasks. While it's hard to assume a boolean system is going to be able to deal with every eventuality, its still likely to have a far higher reliability driving cars than when you introduce human error into the occasion. The same reasoning would apply to my tasks, I think. I have doubts about whether a society too reliant on machines would be stable though, there tends to be a slide towards decadence that is not always healthy. There might be some problems coming at it from that angle, not whether we can trust them but whether we should.
If the technology is reasonably sophisticated, I would trust them to take over many of the mundane tasks. While it's hard to assume a boolean system is going to be able to deal with every eventuality, its still likely to have a far higher reliability driving cars than when you introduce human error into the occasion. The same reasoning would apply to my tasks, I think. I have doubts about whether a society too reliant on machines would be stable though, there tends to be a slide towards decadence that is not always healthy. There might be some problems coming at it from that angle, not whether we can trust them but whether we should.
Page 1 of 1