Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Why human intelligence and AI will co-evolve

nautil.us

Post reply on HN
Print view View original

21–30 of 112 posts

Re: Why human intelligence and AI will co-evolve

#21

Post by randcraw »

jostmey wrote: post #7
eli_gottlieb wrote: post #3

Why do people think they can build anything worthy of the name "AI" by just fucking around with neural networks instead of by specifying exactly what sort of models and exactly what sort of inference constitute "intelligence" in the first place? It's like trying to build an airplane by studying hang-gliders instead of aerodynamics.

The Wright brothers started off by building gliders before they built the first airplane. So why is this approach the wrong one?

Right. Then they built airplanes that don't flap their wings either.

The success of fixed wing aircraft is not an ideal metaphor for the failures of AI research. Nor do the successes of neural nets herald the biomorphic approach, since they don't really resemble the brain very much.

Re: Why human intelligence and AI will co-evolve

#22

Post by 0xdeadbeefbabe »

CmonDev wrote: post #12
0xdeadbeefbabe wrote: post #5

Earlier quoted context omitted.

I thought it already was put to rest by nautil.us http://nautil.us/issue/21/information/the-man-who-tried-to-r... In summary, 1000+ IQ man will discover life is illogical and depressing. He won't do much good when he's so depressed.

Sub-200 IQ men guessing what a 1000+ IQ man would do :).

It's probably safe to say he'd get bullied, but what do I know :).

Re: Why human intelligence and AI will co-evolve

#23

Post by randcraw »

drzaiusapelord wrote: post #14
eli_gottlieb wrote: post #3

Why do people think they can build anything worthy of the name "AI" by just fucking around with neural networks instead of by specifying exactly what sort of models and exactly what sort of inference constitute "intelligence" in the first place? It's like trying to build an airplane by studying hang-gliders instead of aerodynamics.

Because there's no known path to AI and articles like this are pretty much science fiction. Reminds me of that Georges Melies film about going to the moon in a cannon. We didn't have any ideas regarding rockets yet, so cannons it is. I suspect someone will get some level of mega-expert system/simulated intelligence going sooner than later. Maybe something that can handle the types of commands you could give a dog and…

"Get me a Coke..." was accomplished in 1972 by Terry Winograd's SHRDLU and many 3rd gen AI systems since.

NNs will almost certainly succeed where expert systems failed, largely because we now understand that no single monolithic pattern matcher will suffice alone. Any cognitive engine must be composed of many components, each attuned to a different purpose and context. And now we better recognize the huge need for learning, both for initial skill acquisition and for lifelong thereafter.

Minsky's "Society of Mind" is probably a better illustration of how AI will evolve (if not manifest), as well as how it must integrate with our myriad collective needs and personal lives.

Re: Why human intelligence and AI will co-evolve

#24

Post by Retric »

cryoshon wrote: post #18

Pre-genetic-augmentation human intelligence doesn't have to be plain old meat, either-- there's a renaissance of cognitive enhancement occurring right now in the nootropic field, with many different chemical compounds under investigation along with a few less traditional approaches like Thync/tDCS or "brain exercises". As far as eugenic paring of the human gene line toward higher intelligence goes, I wouldn't think a…

I feel people vastly underestimate the difficulty with enhancement. Disease, insanity, obsession, boredom, depression, and on and on it goes. Even twins can have dramatically different physical and mental strengths.

Personally, I suspect this has to do with sci-fi generally hand waving away such things.

Re: Why human intelligence and AI will co-evolve

#25

Post by drzaiusapelord »

randcraw wrote: post #23
drzaiusapelord wrote: post #14

Earlier quoted context omitted.

Because there's no known path to AI and articles like this are pretty much science fiction. Reminds me of that Georges Melies film about going to the moon in a cannon. We didn't have any ideas regarding rockets yet, so cannons it is. I suspect someone will get some level of mega-expert system/simulated intelligence going sooner than later. Maybe something that can handle the types of commands you could give a dog and…

"Get me a Coke..." was accomplished in 1972 by Terry Winograd's SHRDLU and many 3rd gen AI systems since. NNs will almost certainly succeed where expert systems failed, largely because we now understand that no single monolithic pattern matcher will suffice alone. Any cognitive engine must be composed of many components, each attuned to a different purpose and context. And now we better recognize the huge need for le…

Sorry, I meant as a product. Why can't I buy a robot today that can handle "get me a coke" or "fill the dishwasher" or "mow the lawn."

Re: Why human intelligence and AI will co-evolve

#26

Post by woodchuck64 »

> Cognitive engineering, via direct edits to embryonic human DNA, will eventually produce individuals who are well beyond all historical figures in cognitive ability.

Now that would be truly foolhardy. Blindly augmenting intelligence before cognitive moral reasoning is completely understood would be a recipe for incredibly dangerous psychopaths. At least with AI, you don't have a system that has been fine-tuned by eons of evolution to manipulate people at least partly against their own interests.

Re: Why human intelligence and AI will co-evolve

#27

Post by Retric »

0xdeadbeefbabe wrote: post #16
drzaiusapelord wrote: post #14

Earlier quoted context omitted.

Because there's no known path to AI and articles like this are pretty much science fiction. Reminds me of that Georges Melies film about going to the moon in a cannon. We didn't have any ideas regarding rockets yet, so cannons it is. I suspect someone will get some level of mega-expert system/simulated intelligence going sooner than later. Maybe something that can handle the types of commands you could give a dog and…

You must hang around some pretty awesome dogs.

https://www.youtube.com/watch?v=WK4gMe5bx9Y

Re: Why human intelligence and AI will co-evolve

#28

Post by 6d0debc071 »

There is a less happy story in which the attempts at ubermensch are mentally disabled. That they will remain so for quite some time, considering the moral horror potentially involved in the experiments, until the errors we make allow us to gain a fuller understanding of the system sufficient to know which changes are wise to make.

The article seems to take for granted that it is possible to make alterations to a complex system 'improve this set of genetic loci' (paraphrasing), without harming that system. But it gives us no idea what the complexity of improving them is. It may well turn out that the brain is a very fragile piece of spaghetti-coded wet-ware. That if you alter a few of genes even relatively subtly you stand to introduce errors.

Faults in only a few genes have been known to cause undesired phenomena in nature, it would be odd if by purposefully altering them without a reasonable understanding of the system we could avoid those faults.

Machines will probably be 'smarter' in 2050. They will be so next year and this trend has been the case for quite some time. Even if you're prepared to burn through the... consequences... of experimenting with genetic engineering on human cognitive ability, there is no such guarantee for humans.

Re: Why human intelligence and AI will co-evolve

#29

Post by danans »

jraines wrote: post #19
danharaj wrote: post #2

> It is easy to forget that the computer revolution was led by a handful of geniuses: individuals with truly unusual cognitive ability. Great man theory needs to be strangled and put to rest.

Why? This statement is certainly true. You can expand the range of people with any plausible contribution until it's not, but doing so would be an ideological exercise at best. Von Neumann, Shannon, Turing, Weiner, Hopper and many more, on through Wozniak, Torvalds, I could go on and on. There's nothing wrong with calling extraordinary intellect what it is.

I think the objection isn't that those people didn't make huge and transformative contributions to the field.

Rather, it is the tone set by the use of the worshippy word "genius", which has a murky, and totally relative definition, and the phrase "unusual cognitive ability", which implies that their abilities uniquely set these people apart from others who we don't worship in the same way.

Uncounted numbers people have fully understood, and often expanded beyond, the discoveries of Von Neumann, Shannon, Turing etc., since their times, and even more probably had the innate ability to do so, but no access.

Thousands of others have demonstrated the scrappy self-startedness of Wozniak and Torvalds, but without the societal setting and geographic luck that allowed those individuals to succeed.

Given the significant role that one's environment plays in one's success, these people as individuals aren't in themselves unusual. What's unusual is that they were people with the right characteristics, in the right circumstances, and the right support systems.

Basically, extraordinary intellect isn't as unusual or consequential as that sentence from the article implies.

Re: Why human intelligence and AI will co-evolve

#30

Post by minthd »

eli_gottlieb wrote: post #3

Why do people think they can build anything worthy of the name "AI" by just fucking around with neural networks instead of by specifying exactly what sort of models and exactly what sort of inference constitute "intelligence" in the first place? It's like trying to build an airplane by studying hang-gliders instead of aerodynamics.

Because the advancements in the field are interesting and wide ranging: neural networks doing vision, speech, natural language understanding, reasoning and general learning over data.
Post reply on HN