Author Topic: The AI Revolution: The Road to Superintelligence  (Read 13716 times)

Offline Show Me The Exit

  • Anny Roader
  • ****
  • Posts: 489
  • No Mané mo' problems.
Re: The AI Revolution: The Road to Superintelligence
« Reply #40 on: February 2, 2017, 03:17:21 pm »
Thank you too. It is indeed a fascinating topic.

Having been distracted with recent family illness I’m back for more!

I would suggest that the cochlear implant is producing semantic information. But I don't think much rests on this. It's basically the same as any other tool we use. If I use an abacus, the way the beads are arranged in their rows mean something to me. I interpret the abacus as transmitting meaningful information to me even if, as far as the abacus is concerned, it doesn't mean anything at all. It's just an inert collection of atoms that have been arranged in a certain order by an intelligent being that wishes to use it as a tool for meaningful engagement with the world. Nobody would seriously suggest that an abacus is conscious. But apparently a billion abacuses being used at mind-bending speeds at once are somehow supposed to convince me that they are conscious?

Yes – this is a good point, thanks for putting it forward. The thing is though, we, and therefore our consciousness, did not always exist and so at SOME point in our neurodevelopmental history we have evolved from objects that are even simpler than a collection of abaci.

Do we understand how that process has arisen? No.
Did someone direct that process? No (unless you believe in God).
Is it even reliant on semantics? No.

And so this is the hurdle that I’m struggling to overcome. The questions you pose are all interesting but they do not preclude the emergence of a new form of intelligence. If they did then we, ourselves, would not have arisen in the first place.

In an earlier post you say that we do not have to consider that problem since we know that consciousness has already arisen in biological lifeforms (which are clearly made of physical materials). The same process may or may not result in silicone-based forms (for example). Currently, there is no evidence to show that biological forms have any kind of special property that computers do not – apart from the fact that we have not witnessed artificial intelligence yet. But that may just be a matter of time and having the right preconditions in place.

Nevertheless the emergent property is self-organising and the pace of change is exponentially quickening.


EDIT: I wonder whether the source of our divergence in view lies in the idea of artificial intelligence having to be engineered. In one sense it has to be, of course, but in another it can exist through emergence and therefore we may not need to answer the questions you have posed in order to see it.
« Last Edit: February 3, 2017, 10:43:05 am by Show Me The Mané »
Mate, with all due respect, you are an absolute fucking fruit case -MOZ

Seriously mate - You post such utter fucking gash with such conviction it's quite spell binding. - Alan_X

Offline [new username under construction]

  • Poster formerly know as shadowbane. Never lost his head whilst others panicked. Fucking kopite!
  • Legacy Fan
  • ******
  • Posts: 12,284
  • Insert something awesome here!

Offline DHKopper

  • RAWK Supporter
  • Kopite
  • ******
  • Posts: 846
  • We all Live in a Red and White Kop

Offline Raul!

  • No nude LFC topics - Sir Raul la di Dah of Coverpoint - Imminently Female
  • Legacy Fan
  • ******
  • Posts: 16,037
  • My nipples explode with delight
Re: The AI Revolution: The Road to Superintelligence
« Reply #43 on: March 6, 2018, 05:22:48 am »
For anyone interested in the topic, and I sincerely believe that everyone should be, here is an absolutely fascinating conversation between two of the most intelligent people on the planet that I have come across. Warning, It is pessimistic, without being dystopic and the reasoning is highly rational and more than a bit plausible. If you have the time, these will be two plus of the more illuminating hours you will spend this week, month or year.

Can't recommend this highly enough.

https://samharris.org/podcasts/116-ai-racing-toward-brink/

Offline The Gulleysucker

  • RAWK's very own spinached up Popeye. Transfer Board Veteran 5 Stars.
  • RAWK Remembers
  • Legacy Fan
  • ******
  • Posts: 11,496
  • An Indolent Sybarite
Re: The AI Revolution: The Road to Superintelligence
« Reply #44 on: March 6, 2018, 10:42:19 am »


Cheers for that, I'll give it a listen.
I don't do polite so fuck yoursalf with your stupid accusations...

Right you fuckwit I will show you why you are talking out of your fat arse...

Mutton Geoff (Obviously a real nice guy)

Offline kavah

  • the Blacksmith. Definitely NOT from Blackpool!
  • RAWK Supporter
  • Legacy Fan
  • ******
  • Posts: 19,662
  • We all Live in a Red and White Kop
Re: The AI Revolution: The Road to Superintelligence
« Reply #45 on: March 6, 2018, 11:43:46 am »
For anyone interested in the topic, and I sincerely believe that everyone should be, here is an absolutely fascinating conversation between two of the most intelligent people on the planet that I have come across. Warning, It is pessimistic, without being dystopic and the reasoning is highly rational and more than a bit plausible. If you have the time, these will be two plus of the more illuminating hours you will spend this week, month or year.

Can't recommend this highly enough.

https://samharris.org/podcasts/116-ai-racing-toward-brink/

nice one

Offline The Gulleysucker

  • RAWK's very own spinached up Popeye. Transfer Board Veteran 5 Stars.
  • RAWK Remembers
  • Legacy Fan
  • ******
  • Posts: 11,496
  • An Indolent Sybarite
Re: The AI Revolution: The Road to Superintelligence
« Reply #46 on: August 30, 2018, 07:56:20 am »
Not directly AI, but there's a long'ish but cracking read in todays Grauniad here regarding the dangers that can arise in self learning and the correspondingly increasing difficulties in being able to predict the end results of such behaviour, with some interesting embedded links also within it too.

‘What’s the worst thing that could happen in this algorithm once it starts interacting with others?’ The problem is we don’t even have a word for this concept, much less a science to study it.”

He pauses for moment, trying to wrap his brain around the problem.

“The thing is, optimizing is all about either maximizing or minimizing something, which in computer terms are the same. So what is the opposite of an optimization, ie the least optimal case, and how do we identify and measure it? The question we need to ask, which we never do, is: ‘What’s the most extreme possible behavior in a system I thought I was optimizing?’”

Another brief silence ends with a hint of surprise in his voice.

“Basically, we need a new science,” he says.


For those interested in such things, grab a coffee or tea, sit down, and and have a read.
I don't do polite so fuck yoursalf with your stupid accusations...

Right you fuckwit I will show you why you are talking out of your fat arse...

Mutton Geoff (Obviously a real nice guy)

Online SamAteTheRedAcid

  • Currently facing issues around potty training. All help appreciated.
  • RAWK Supporter
  • Legacy Fan
  • ******
  • Posts: 22,203
Re: The AI Revolution: The Road to Superintelligence
« Reply #47 on: August 30, 2018, 11:30:00 am »
Very interesting read that Gulley. I'm not sure I entirely understood it all but a few bits will stick in my head. Its a frightening thought that we are now building algorithms that could go well beyond our control. The car not being able to discern a woman on her bike with shopping bags is exactly the kind of problem they have - humans are still unpredictable in that way, and the computer in that instance made the wrong assumption.
get thee to the library before the c*nts close it down

we are a bunch of twats commenting on a website.

Offline stoopid yank

  • Legacy Fan
  • ******
  • Posts: 2,113
  • Bird is the Word
Re: The AI Revolution: The Road to Superintelligence
« Reply #48 on: August 30, 2018, 03:07:20 pm »
When the AI stuff comes up I always think of Stephen Hawking and his story/joke/warning. It was something along the lines of:


"When the first AI is built, it will be asked 'Is there a god'

The AI will respond 'There is now' "

Haunting.



I don't always listen to Black Sabbath, but when I do, so do the neighbors.

Offline The Gulleysucker

  • RAWK's very own spinached up Popeye. Transfer Board Veteran 5 Stars.
  • RAWK Remembers
  • Legacy Fan
  • ******
  • Posts: 11,496
  • An Indolent Sybarite
Re: The AI Revolution: The Road to Superintelligence
« Reply #49 on: August 31, 2018, 01:48:30 pm »
Very interesting read that Gulley. I'm not sure I entirely understood it all but a few bits will stick in my head. Its a frightening thought that we are now building algorithms that could go well beyond our control. The car not being able to discern a woman on her bike with shopping bags is exactly the kind of problem they have - humans are still unpredictable in that way, and the computer in that instance made the wrong assumption.

It is interesting stuff.

I think the problem we face could be described in a simplified way as that over the last 15 years or so, there has been a move towards what is called service orientated computing.

This has been  an evolved solution developed largely to mitigate the problems of years gone by in IT and to take into account developments in technology that have allowed the once fanciful dreams of distributed computing to finally become reality and start to come of age.

But it's possibly all becoming a bit wild west in some respects and I think that's the jist of the article.

For example a service ( a process offering some clearly defined functionality in a computer and with a standardised way of conversation) was simply something that a client (an end user represented by a process or another service in often a different computer) connected to in order to get some required functionality performed.

These things would traditionally be clear cut business or logic processes, ie return a current share price, calculate VAT etc, calculate torsional strength, and easily defined in code by instructions to read this, do this to it and then output the results type of work, and also all easily managed, delineated and very predictable and invariant in their behaviour and with modern programming languages, very safely coded.

But over the last 10 years, these distributed services, often from 3rd party external providers, have sometimes independently evolved and become more sophisticated and can now vary their behaviour and responses by the implementation of self learning due to the incredibly powerful machines and their cheap processing power that we now have at our disposal.

That makes it more difficult to predict with any reliability the complete behaviour of a system in dare I say a holistic way. It's moving from deterministic to non-deterministic in predictability..

Now possibly 99.999% of the time it will still behave as expected and all will be well, no children will be harmed, but it's that 0.001% that should concern us and what we just don't know or have difficulty in knowing as these distributed services, or the behaviour of the components within, evolve through self learning algorithms.

As an analogy, I think there are indications we are possibly reaching the stage when these distributed services are behaving, or have been given the ability to behave, almost like independent countries in that while their name may remain the same, they can have changes of government and corresponding changes in foreign policies and thus their relations with other countries. We know how fraught that can be on the international stage with misunderstandings of intent.

So I'd go so far as to suggest that examination of the foreign policy behaviours and diplomacy of nations might possibly produce a basis of a model for the interactions of these new self learning systems though I also believe we should also examine the effect on international relations that employment of 'winning' strategies may result in.

You see I'm of a mind that if we always code 'winning' ( ie agression or perceived agressive behaviour) as the ultimate objective of any autonomous learning algorithm, while such behaviour would be attractive perhaps within the financial trading world and is employed, their will be trouble if we don't provide the means of scoping it, and for us all.

So we perhaps also need a way to codify such aggression or winning or whatever we may call it, but also test it with reliability for out of bounds behaviour and home in on that 0.001% and remove it.

But I'm waffling a bit here,  so I'll go away and give it some more thought.
I don't do polite so fuck yoursalf with your stupid accusations...

Right you fuckwit I will show you why you are talking out of your fat arse...

Mutton Geoff (Obviously a real nice guy)

Offline The Gulleysucker

  • RAWK's very own spinached up Popeye. Transfer Board Veteran 5 Stars.
  • RAWK Remembers
  • Legacy Fan
  • ******
  • Posts: 11,496
  • An Indolent Sybarite
Re: The AI Revolution: The Road to Superintelligence
« Reply #50 on: August 31, 2018, 02:04:19 pm »
Just a quick afterthought.

I recall seeing some research documents back in the 80's that rolled out of the back of Reagan's Starwars programme.

Since the proposed Death Stars (H bombs in orbit that had their x-ray emission focussed through glass rods during the 3 millionaths of a second before destruction and could emit simply mindboggling  power level X Ray laser beams targetted at missiles in flight) would be totally autonomous, there was naturally concern if they made a mistake and triggered their activity unannounced due to unforseen behaviour.

The papers I read were outlining initial stabs at proposing a language for mathematical proof of software reliability in order to validate such things.

I haven't looked at such stuff in years, but I expect a resergence in such areas might be fruitful if not already underway.

« Last Edit: September 1, 2018, 06:47:26 pm by The Gulleysucker »
I don't do polite so fuck yoursalf with your stupid accusations...

Right you fuckwit I will show you why you are talking out of your fat arse...

Mutton Geoff (Obviously a real nice guy)

Offline The Gulleysucker

  • RAWK's very own spinached up Popeye. Transfer Board Veteran 5 Stars.
  • RAWK Remembers
  • Legacy Fan
  • ******
  • Posts: 11,496
  • An Indolent Sybarite
Re: The AI Revolution: The Road to Superintelligence
« Reply #51 on: October 11, 2018, 09:53:26 pm »
An interesting article, short and a bit lacking in real technical substance, but worth a glance through to see what's perhaps being considered as far as the Pentagon is concerned...

https://www.theguardian.com/commentisfree/2018/oct/11/war-jedi-algorithmic-warfare-us-military
I don't do polite so fuck yoursalf with your stupid accusations...

Right you fuckwit I will show you why you are talking out of your fat arse...

Mutton Geoff (Obviously a real nice guy)

Offline thejbs

  • well-focussed, deffo not at all bias......ed
  • Legacy Fan
  • ******
  • Posts: 8,617
Re: The AI Revolution: The Road to Superintelligence
« Reply #52 on: June 1, 2019, 01:30:40 am »
Fascinating AI attempts at creating people.

https://thispersondoesnotexist.com/

Offline Andy @ Allerton!

  • Missing an asterisk - no, wait sorry, that's his rusty starfish..... RAWK Apple fanboy. Hedley Lamarr's bestest mate. Has done nothing incredible ever.
  • RAWK Supporter
  • Legacy Fan
  • ******
  • Posts: 73,319
  • Asterisks baby!
Re: The AI Revolution: The Road to Superintelligence
« Reply #53 on: June 3, 2019, 04:37:12 pm »
Can these do great AI on a wet and windy Wednesday night in Stoke?
Quote from: tubby on Today at 12:45:53 pm

They both went in high, that's factually correct, both tried to play the ball at height.  Doku with his foot, Mac Allister with his chest.

Offline [new username under construction]

  • Poster formerly know as shadowbane. Never lost his head whilst others panicked. Fucking kopite!
  • Legacy Fan
  • ******
  • Posts: 12,284
  • Insert something awesome here!
Re: The AI Revolution: The Road to Superintelligence
« Reply #54 on: October 31, 2019, 01:39:48 pm »
Love these little guys :D

https://openai.com/blog/emergent-tool-use/

Popped up in a video on Quantum Supremacy and AI

<a href="https://www.youtube.com/v/92Ntk4niqPo" target="_blank" rel="noopener noreferrer" class="bbc_link bbc_flash_disabled new_win">https://www.youtube.com/v/92Ntk4niqPo</a>

Offline Andy @ Allerton!

  • Missing an asterisk - no, wait sorry, that's his rusty starfish..... RAWK Apple fanboy. Hedley Lamarr's bestest mate. Has done nothing incredible ever.
  • RAWK Supporter
  • Legacy Fan
  • ******
  • Posts: 73,319
  • Asterisks baby!
Re: The AI Revolution: The Road to Superintelligence
« Reply #55 on: April 27, 2020, 11:51:03 pm »
AI is still rubbish.

It's all very well learning definitive skillsets with defined paramters like Go! or Chess or Draughts.


For me, an AI would be actually scary/remotely good if you could give it Zork, Planetfall or The Pawn when it has no pre-programming of what a text-based IF title is.

It would have to sort out working how to type into the interface, recognise and learn and understand the responses, work out the words (I'd let it have an English dictionary) and then work out what it has to do, objects it has to interact with, puzzles it has to solve, constraints offered in the output and lateral thinking to arrive at answers.

As clever and amazing as AI is, as far as I'm aware, there is no AI on the planet that could even beat Zork without being pre-programmed to do so.



I was randomly thinking this today and then came across this article, which is pretty interesting:   http://www.spagmag.org/issue-64/solve-zork-teaching-an-ai-to-play-parser-if/


Here is another view on it:  https://www.theregister.co.uk/2018/07/16/microsofts_textworld_gives_ais_a_zorklike_challenge/

And also this:   https://www.independent.co.uk/life-style/gadgets-and-tech/teaching-computers-doom-ai-zork-colossal-cave-programmers-a8296701.html
« Last Edit: April 27, 2020, 11:56:03 pm by Andy @ Allerton »
Quote from: tubby on Today at 12:45:53 pm

They both went in high, that's factually correct, both tried to play the ball at height.  Doku with his foot, Mac Allister with his chest.

Offline Zlen

  • Suspicious of systems. But getting lots.
  • RAWK Scribe
  • Legacy Fan
  • ******
  • Posts: 18,791
Re: The AI Revolution: The Road to Superintelligence
« Reply #56 on: April 28, 2020, 08:17:32 am »
Nice, thanks for sharing.