5
$\begingroup$

Given the advantage AI already has over human intelligence, one could imagine a relatively weak strong-AI (barely human intelligence) still outperforming a segment of the human scientist population in terms of scientific discoveries per year (or hour).

Will AIs be doing most of the science in 50 years?

$\endgroup$
  • 3
    $\begingroup$ As of now you can come up with plausible reasoning for both for and against the claim, but the reality as of now is that we don't know $\endgroup$ – Ankur Nov 23 '16 at 5:28
  • 1
    $\begingroup$ What I wonder about is how much of a factor is creative thinking in scientific advancement, particularly with the great leaps imagination that seem to underpin major breakthroughs, and will machines be able to easily replicate that. I suspect the game trees on "great leaps of the imagination" would be rather daunting, even to a powerful, future computer. $\endgroup$ – DukeZhou Nov 24 '16 at 20:01
3
$\begingroup$

According to Ray Kurzweil, a prominent AI researcher, yes. In his book The Singularity is Near he predicts that AIs will take over developing other AIs in about 30 years, after which human intelligence will become marginalised.

$\endgroup$
0
$\begingroup$

I don't think so, it is not the first but actually the third wave of neural networks. It's doing better than earlier two as we have much more amount of data as well as computational power now. Take a look at this video ..... According to Douglas Adams’s famous “Hitchhiker’s Guide to the Galaxy” after 7.5 millions years of work the “Deep Thought” computer categorically found out that 42 is the “Answer to the Ultimate Question of Life, the Universe, and Everything” (although unfortunately, no one knows exactly what that question was).

$\endgroup$
0
$\begingroup$

Current Computing relates 0 or 1 to another 0 or 1 with layers upon layers of building blocks built on this relationship.

In future AI will relate A to B, be they numbers, patterns of coded instructions or some other form of more complicated constructs at a hardware (or closer to it) level than is currently possible and, due to the inherent perfect recall and potentially massive memory storage of AI they will most definitely be bringing together and relating a vastly more broad and organised collection of knowledge than it is possible for any human to even contemplate consciously. There are some ideas that would argue that point; universal mind, spiritual revelation and morphic resonance which I do personally agree with to some degree and can imagine being particularly difficult to represent in a computational format.

Pattern spotting and relating, organising and computing potentials... computers are already better at all these things than most people try to be. It will not be long, i think, before they can "invent" something "new".

There are already attempts at AI in specialised fields of knowledge, human speech, various games, medical diagnostics and learning etc. It will be when these specialised AIs can "compare notes" about the various methodologies that have been the most productive or rewarding, in whichever form these take for them, and accordingly update their own ontologies that the true explosion of "Intelligence" will occur.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.