Re:John Carmack: I’m going to work on artificial general int...
by K0SM0S on Wednesday November 13, 2019 @07:12PM
(#21531311)
I'll say this: I'd prefer if the brightest minds approached the matter from the "AI safety" angle (a sub field concerned with building not just AI but "safe" AI, ie that we can control or understand in a practical manner).
Because that's really where the line of human history will be drawn if AGI and above becomes real. AI safety, how advanced we are in it, will directly map to civilization's progress or endangerment as a result of AI.
Edit: this is already true with regards to "psychological safety" from undue influence or outright manipulation with motive (usually financial) by current "ANI" algorithms (newsfeeds, "recommendations", ads, etc). It's a real topic that reduces to human psychological freedom, freewill. It's a BIG topic.
Because that's really where the line of human history will be drawn if AGI and above becomes real. AI safety, how advanced we are in it, will directly map to civilization's progress or endangerment as a result of AI.
Edit: this is already true with regards to "psychological safety" from undue influence or outright manipulation with motive (usually financial) by current "ANI" algorithms (newsfeeds, "recommendations", ads, etc). It's a real topic that reduces to human psychological freedom, freewill. It's a BIG topic.