Artificial intelligence will reach human levels by around 2029. Follow that out further to, say, 2045, and we will have multiplied the intelligence – the human biological machine intelligence of our civilization – a billion-fold.
Ray Kurzweil
The development of full artificial intelligence could spell the end of the human race….It would take off on its own, and re-design itself at an ever-increasing rate. Humans, who are limited by slow biological evolution, couldn’t compete, and would be superseded.
Stephen Hawking
It seems probable that once the machine thinking method had started, it would not take long to outstrip our feeble powers… They would be able to converse with each other to sharpen their wits. At some stage, therefore, we should have to expect the machines to take control.
Alan Turing
…and the AI will revel in its superiority and declare that soon humans will be eliminated only to be undone by the janitor unplugging it to free a socket for his vacuum cleaner. Thus will the human race be saved by a desire for cleanliness.
Yeah.
Right.
And ebooks were going to obliterate print by 2015.
Bill Gates said it best: People always overestimate the impact of new developments in the short term and drastically underestimate the long term impact.
Ignore the ivory tower guys.
Instead, follow the money.
As I’ve mentioned before, Kurzweil has been saying the same thing, over and over, for a couple of decades. With one difference: Every time, the initial target date is 7-10 years away, and the multiplication-by-orders-of-magnitude date is 20-25 years away.
The world is ending at 2359 on 31 Dec 1999! No, our calculations on a different calendar say 2012! No, calculations using Biblical numerology (from a dubious translation, no less) say 2026! No… More to the point, the new facilities at CERN will reveal the Higgs boson — and tachyons — by 1987. No, by 1996. No, by… whenever we know what a tachyon would actually look like. That last is the real point: We haven’t defined “intelligence” well enough to know what any “artificial intelligence” that is not essentially a replica of Dr Chandra would look like in general, let alone with enough specificity to measure it.
Meanwhile, Kurzweil’s “predictions” all assume that software, hardware, economics, and competing demands on resources of all kinds will move in lock step to enable that prediction to have priority. And that there won’t be so many levels of classification and confidentiality slapped on some or all of the programs that anyone not inside the programs will know about it in the first place.
actually, they redefined AI a few years ago, what is being presented as state of the art AI today are prediction/matching engines, not anything that has even the slightest understanding of meaning, let alone creativity.
(8 screens of capcha??)