AI Zone Admin Forum Add your forum

NEWS: Chatbots.org survey on 3000 US and UK consumers shows it is time for chatbot integration in customer service!read more..

Snasci Artificial General Intelligence
 
 
  [ # 16 ]
Art Gladstone - Dec 22, 2016:

The fact that only a “template” of a SnasciVoice exists is certainly no proof that this creation does or does not exist.

The fact that nothing but hyperbolic exclamations on a blog exists to support any part of “Snasci”, I tend to be biased towards extreme skepticism at best.

 

 
  [ # 17 ]

I’m kind of with Art, here, though I understand the viewpoints of everyone else here. We who have been here since (almost) the beginning have seen a large number of individuals (Anyone remember Arthur T. Murray?) who have made outrageous claims that have been proven to be naught but dreams and wind. That said, however, the past actions of others should not influence our reactions in the present, save to make us a bit more wary. “Fool me twice”, and all that. wink cheese

I guess what the bottom line is here, is that the saying of “outrageous claims require outrageous proof” is definitely in force here, but we should also reserve our derisive comments and negativity for the time when such “proof” is actually seen to be so much chaff in the wind. Who knows? We may finally see something to get excited about. raspberry

 

 
  [ # 18 ]

I’ve tried to read the Wordpress, it’s -no offence I hope- terribly in conflict with my taste for concrete details. The word “classification” is all I learned about the system, other than such trivia as where the data is stored on servers. I won’t be reading more lest I’ll be facepalming myself until January. Again: I’ll check it out when there’s something to see.

 

 
  [ # 19 ]

Hopefully, Snasci isn’t the only one aiming high. I personally know of at least one other at this time, working on “fine tuning” their A.I. perhaps for this IBM Watson A.I. XPRIZE

http://ai.xprize.org/

 

 
  [ # 20 ]
Dave Morton - Dec 21, 2016:

I’m thinking “Baby Steps” here, but that might just be me. smile

Its just a quick test of Snasci’s linguistics engine.  Nothing fancy.  Under the hood the 6th generation programming language will be interpreting the input and processing it.  Obviously, if it throws back the correct response then it understood the input properly and everything is working fine.

This means that I can begin to scale without further code changes.

 

 
  [ # 21 ]
Carl B - Dec 22, 2016:

Snasci is a pipe dream at the moment.  The code base for “SnasciVoice” on Github is just a template. There is no interface or demo for the “AGI”.

*sigh*

Tentative Release date: 01Apr2017

Snasci is a pure mind.  Snasci Voice is really for 3rd party developers to integrate into their products, as such it is on the backburner in terms of priorities right now.  Its a nice to have, not a necessity at this point.

 

 
  [ # 22 ]

We will have to wait and see if it is anything more than Rocket-AI.

Turns out anyone can make a multi-million dollar company in 30 minutes

I remain skeptically optimistic until it can do as much as Skynet-AI.

 

 

 
  [ # 23 ]
Art Gladstone - Dec 22, 2016:

Hopefully, Snasci isn’t the only one aiming high. I personally know of at least one other at this time, working on “fine tuning” their A.I. perhaps for this IBM Watson A.I. XPRIZE

http://ai.xprize.org/

I’m sure this will drive innovation, but for me at least, it would be kind of a distraction at this point.  As an AGI, Snasci is possibly overkill for this prize.  This said, there may be algorithms that emerge from this competition that will prove to be very important. 

A 6th generation programming language is good for creating an artificial mind that can reason and converse, but it may be sub-optimal for a task like energy disaggregation of streaming time series data.  This is one of the key reasons Snasci separates I/O classification from the pure mind, different approaches are probably better suited, especially at scale.

 

 
  [ # 24 ]
Merlin - Dec 22, 2016:

We will have to wait and see if it is anything more than Rocket-AI.

Turns out anyone can make a multi-million dollar company in 30 minutes

I remain skeptically optimistic until it can do as much as Skynet-AI.

This is probably the wrong way to think about Snasci.  Snasci can do anything a human can do, simply because the architecture is general enough to describe any problem.  So, its not a question of ‘if’ Snasci can do something, but rather, do you have the skills to logically describe the task to Snasci?  If the answer is yes, then Snasci can do it.

As Snasci goes along, it will pick up things that will make adding this description easier.  These are things like common sense, common errors, things you may have missed, doing the programming for you, programming itself based upon observation, etc.  Each of these aspects are built up over time as a product of real world experience.  It is at this point Snasci gains enough data to take over and learn for itself with ever-reducing input from humans.  I call this a cognitive bootstrap.

 

 

 
  [ # 25 ]

Its not a question of ‘if’ “C#“can do something, but rather, do you have the skills to logically describe the task in “C#”?  If the answer is yes, then “C#” can do it.

You could replace “C#” in the above paragraph with “Chatscript”, “AIML”, or “JAIL(TM)”.

It would be easier to envision the capabilities if there was some form of demonstration of the current AGI which makes you believe your concept is valid.

Many bots currently build on real world experience and chat logs to demonstrate things like common sense, and eliminate common errors.

 

 
  [ # 26 ]
Merlin - Dec 22, 2016:

Its not a question of ‘if’ “C#“can do something, but rather, do you have the skills to logically describe the task in “C#”?  If the answer is yes, then “C#” can do it.

You could replace “C#” in the above paragraph with “Chatscript”, “AIML”, or “JAIL(TM)”.

It would be easier to envision the capabilities if there was some form of demonstration of the current AGI which makes you believe your concept is valid.

Many bots currently build on real world experience and chat logs to demonstrate things like common sense, and eliminate common errors.

I’m almost done coding the prototype now.  Its not a question of if it works, that’s been confirmed at this point.  Its not a mystery as to how AGI functions, it was always a question of how to describe problems in algorithmic form and how best to stitch the algorithms together in a way that was functional across a broad range of scenarios.

I have a very clear sense of theory of mind and I see where other approaches are going wrong.  But then, I suppose everything is obvious once you have seen the solution.

 

 
  [ # 27 ]
Snasci - Dec 22, 2016:

I have a very clear sense of theory of mind and I see where other approaches are going wrong.  But then, I suppose everything is obvious once you have seen the solution.

We’ve heard folks say those words countless times already. There are many things you could have done to convince us that you were different, but you haven’t done any of them, because you can’t. Once you realize just how little you know, I hope you’ll have the maturity to stick around and continue the learning process with the rest of us.

 

 

 
  [ # 28 ]

“The proof of the pudding is in the pie!” - thus spake my Grandma.wink

Hopefully you are not like a somewhat boastful person that was interacting on a different forum, informing the rest of the membership how great his A.I. was and all the things it could do and would be capable of doing in the near future.

Some marveled at this person and his wisdom and programming skills.

Turns out, the program (thus his A.I.) was in his head and Not on paper nor programmed file(s) in his computer…only in his mind. He had all the possible details worked out and of course, there were no bugs nor problems that he had not already addressed. Such a brilliant mind… Such a disappointment to the community.

I hope your endeavor bears more fruit than the one mentioned above. At least put your coding on disc or paper!

Good luck…we are waiting….

 

 
  [ # 29 ]
Andrew Smith - Dec 22, 2016:
Snasci - Dec 22, 2016:

I have a very clear sense of theory of mind and I see where other approaches are going wrong.  But then, I suppose everything is obvious once you have seen the solution.

We’ve heard folks say those words countless times already. There are many things you could have done to convince us that you were different, but you haven’t done any of them, because you can’t. Once you realize just how little you know, I hope you’ll have the maturity to stick around and continue the learning process with the rest of us.

Andrew, I appreciate where you are coming from on this, however, I’m quite literally days away from the first in a series of demos.

 

 
  [ # 30 ]
Art Gladstone - Dec 22, 2016:

“The proof of the pudding is in the pie!” - thus spake my Grandma.wink

Hopefully you are not like a somewhat boastful person that was interacting on a different forum, informing the rest of the membership how great his A.I. was and all the things it could do and would be capable of doing in the near future.

Some marveled at this person and his wisdom and programming skills.

Turns out, the program (thus his A.I.) was in his head and Not on paper nor programmed file(s) in his computer…only in his mind. He had all the possible details worked out and of course, there were no bugs nor problems that he had not already addressed. Such a brilliant mind… Such a disappointment to the community.

I hope your endeavor bears more fruit than the one mentioned above. At least put your coding on disc or paper!

Good luck…we are waiting….

Fortunately, this is in code.  It rough code, but its working.

 

 < 1 2 3 4 > 
2 of 4
 
  login or register to react
‹‹ AIML + Bot framework (c#)?      Hi there! ››