3 ms·
Existential risk around AI? Really? That's just multiplying through by -1 the typical nerd wet dream about the Singularity. I can't even keep Rails, Apache, o
by angersock 14y ago
Existential risk around AI? Really?
That's just multiplying through by -1 the typical nerd wet dream about the Singularity.
I can't even keep Rails, Apache, or git from occasionally shitting the bed; I am not worried about AI just yet.
- ubercore 14y agoWhile I completely and wholeheartedly agree with your sentiment, and think fear-mongering is just that, the interesting thought experiment is thinking what technology would look like for something that _didn't_ shit the bed. Or when it did, it fixed itself. That, plus a little bit of intelligence, could (would?) be scary.
- mej10 14y agoYes, existential risk around AI. Really. You can look at it that way, but I think it is currently far more likely than the "typical nerd wet dream about the Singularity". Your last sentence is a complete non sequitur to me. You are comparing the output of a large, AI focused, for profit organization to an individual of unknown skill using some very specific open source tools. I am not losing sleep over it, but I definitely think about it from time to time.
- keefe 14y agomej10 you sound like someone singinst associated : ] also not being able to keep git etc. running is some evidence about skill level
- bitwize 14y agoExistential risk around AI? Really? I don't think it unreasonable to brace for the case where some Skynet wannabe launches all nukes not because of a carefully considered decision motivated by hatred for humanity, but simply because it shat the bed.
- theorique 14y agoThat's just multiplying through by -1 the typical nerd wet dream about the Singularity. Well said. What is it that AI is supposed to do that's so scary, anyway?
- Jach 14y agoWhat scary things do humans do, to each other and to lesser species? What amazing things do humans do? What concept lets humans do anything noteworthy at all? Here's a nice essay on the power of intelligence: http://yudkowsky.net/singularity/power http://yudkowsky.net/singularity/power
- taligent 14y agoSince when does intelligence = human intelligence ? You would think an intelligent person would know this.
- Jach 14y agoIt doesn't. Hence the curiosities and concerns about general AI, non-human intelligence, hopefully created by us humans, the first intelligences intelligent enough to design their successor instead of letting a weaker non-intelligent design process such as evolution happen upon it by chance. The site I linked to has some good introductory material to the related concepts. (http://yudkowsky.net/singularity/intro http://yudkowsky.net/singularity/intro is short and to the point.) http://singularity.org/research/ http://singularity.org/research/ lists some important texts to read, and following up on the authors will lead to more. (An author I will also recommend following for his practical pursuits is Ben Goertzel. Lastly, I think this is a good criticism of the Singularity Institute's general mode of operating: http://kruel.co/2012/11/03/what-i-would-like-the-singularity-institute-to-publish/ http://kruel.co/2012/11/03/what-i-would-like-the-singularity... But you should probably read that one last, if you intend to read anything at all beyond the simple introduction.)
- div 14y agoDepending on how you define AI, anything between nothing and everything.