You are viewing a single comment's thread.

view the rest of the comments →

0
1

[–] rwbj 0 points 1 point (+1|-0) ago  (edited ago)

I'm conflicted about Ray. His past successes alone deserve unbreakable respect and so he gets that. However, I think his predictions for the ~2030s are being somewhat unrealistically raced forward. I think that is him, probably subconsciously, working to reconcile his own mortality. Many of his 'post-human' predictions are kind of out coming out of nowhere and coincide almost perfectly, timeline wise, with him being able to utilize them just prior to his death. Of course I hope I'm wrong, but I don't think I am.

It's kind of like "Yeah, all cars will be self driving 15 years." A reasonable futuristic expectation but then he also adds, "Oh and all disease will be eliminated and we'll be able to integrate and even scan our brain into machines capable of simulating human level intelligence based on a scan of human tissue." We haven't even begun to touch the mind-machine interface. To have it completely worked out in 15 years - I don't know. I don't see any reasonable justification for that. Even if you assume increasingly exponential technological progress - we literally are starting from 0. Zapping the brain's surface and observing it does this or that is no more understanding than a person who realizes that if he slams his hands on his keyboard long enough that the computer will start beeping can claim to understand said computer.

--

edit To clarify I can completely imagine human level AI developed within the next 15 years. And that would last about a micro-second before we reach post-human level intelligence. The issue I have is not with the AI which one can only imagine where we're going, but with his linking AI and the human mind. There's just no basis in reality for 'transhumanist technology' here.

0
1

[–] tame 0 points 1 point (+1|-0) ago 

I kind of agree, Kurzweil's done some awesome stuff but I think his recent work (especially his predictions) are more wish-fulfillment beliefs than solid science. We are making headway to where he thinks we're going but we're not so far up the curve towards the singularity as I think he's hoping we are.

0
0

[–] ideasware [S] ago 

0
2

[–] rwbj 0 points 2 points (+2|-0) ago 

The first article is the best example of the problem. That technology has absolutely 0 to do with breaking the mind machine interface. They're effectively looking to measure the brain's output in response to various stimuli and create a correlationary mapping. This, for instance, is the exact same science used in lie detector machines. The only difference is you're directly measuring brain activity as opposed to pulse, temperature, respiration, etc.

The big issue is causation vs correlation. With correlations you can create quite a large number of predictive tools or useful interfaces. But without causation you'll perpetually be an observer and not a 'player.'. Take for instance depression. What causes depression? It's a trick question since the real answer is we have absolutely no clue. If you thought the answer had to do with serotonin, you'd be incorrect. What we do know is that there is a correlation between SSRIs and positive changes in regards to depression, but we have no clue what or why this is. It may not even have anything to do with seratonin but instead be another change with the brain and overall physiology that the SSRIs trigger. This is the difference between causation and correlation. And causation is an area where we are still literally at 0 in terms of knowledge as it relates to the brain.