3 ms·
I would argue the risks involved in creating a "real" AI are present regardless. Not to re-link, but this example is even more apropos to your question than my
by ericb 14y ago
I would argue the risks involved in creating a "real" AI are present regardless. Not to re-link, but this example is even more apropos to your question than my other answer in this post.
http://wiki.lesswrong.com/wiki/Paperclip_maximizer http://wiki.lesswrong.com/wiki/Paperclip_maximizer
- taligent 14y agoAI is just smarter software at the end of the day. The risks are still far greater for a simple control system used in a Boeing 777 or heart pacemaker than for an AI search engine. It's all about Risk Management 101: Risk = Impact x Probability.
- ericb 14y agoRight, when the impact is calamity, and the probability is high, risk is astronomical. Which is why I'm bothering to make this point in the first place. See the other arguments in this thread as to why, with an intelligent AI, what it starts out controlling is unlikely to have any bearing on what it is capable of gaining control of... In risk management terms this is about as serious as it gets.