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..

AiMind
 
 
  [ # 121 ]

Steve Worswick (a.k.a “squarebear”) posted a transcript
both here upthread and also where I noticed it first at

http://www.ai-forum.org/topic.asp?forum_id=1&topic_id=73371

which is also where I posted an immediate, initial response.
Afterwards I was thinking about why the AiMind.html gave
no pertinent response when Steve asked, “what does fido have”.
The reason was that the tabula-rasa AI Mind had not made an
inference from the singular-subject premise “a dog has a tail”
and therefore the AI may have tried to generate a thought
starting with “FIDO HAS…” but there was no direct object
available, so the incipient thought got aborted for lack of
a direct-object noun like “TAIL” to complete the thought.
Then the AI lamely said, “I AM ANDRU” in its aporia.

Rejoining the discussion here at Chatbots.org, I once
again called up the JavaScript AI to ask it Steve’s question:

********** START OF AI MIND TRANSCRIPT ************
Time = 778; thinking with NLP loop #2; ReJuvenate = 0
AiMind version 14apr13A on Mon May 27 09:33:12 PDT 2013

Robot: I HELP KIDS
Human:

Robot:
Human: dogs have tails

Robot: I AM PERSON
Human:

Robot: WHAT ARE TAILS
Human:

Robot:
Human: fido is a dog

Robot: DOES FIDO HAVE TAILS
Human: yes

Robot: FIDO IS DOG
Human: what does fido have

Robot: FIDO HAVES TAILS
Human:

Robot: WHAT IS A FIDO
Human:
*********** END OF AI MIND TRANSCRIPT *************

I did not switch to Transcript mode quickly enough,
so the AI thought its own thoughts in Tutorial mode.
However, once the inference was confirmed with a
“yes” answer, the AI knew that Fido, being a dog,
has tails (I forgot to use “tail” in the singular), and
so the AI could give a better answer to Steve’s query.

With Vincent Gilbert’s transcript posted above,
using “What’s your name” as a name is confusing.
It is pretty cool that the chatbot remembers that
at least one dog (or “woofie”) has four legs,
so the chatbot program is obviously advanced.
Congratulations, Vincent Gilbert!

Are there any other so advanced AI programs?

 

 

 
  [ # 122 ]

This program can also take “x has a y”, “z is a x” and infer from it.

http://www.pandorabots.com/pandora/talk?botid=9c5043571e34414f

Human: tigers have claws
testbot: That’s cool to know.

Human: tony is a tiger
testbot: So does tony have claws?

Human: what does tony have?
testbot: I assume tony has claws?

 

 
  [ # 123 ]

I think it is a shame Athur T Murray is approaching this from the point of view of already having a great AI. It is only the intial building blocks, but interesting blocks nonetheless. Frankly it’s good to see someone approaching this differently.

However, my recommendation to Arthur is to stop using the techno-babble, approach this for what it is (a very early stage concept that has yet to show it’s potential) and make it a project that could inspire others to get involved and perhaps make it into something more.

 

 
  [ # 124 ]
MikeA - Jun 10, 2013:

It is only the initial building blocks, but interesting blocks nonetheless. Frankly it’s good to see someone approaching this differently.

http://ai.neocities.org/AiSteps.html—is the “initial building blocks”.

http://www.nlg-wiki.org/systems/User:Arthur_T._Murray is my new wiki-page.

http://www.nlg-wiki.org/systems/Mind—Natural Language Generation wiki.

http://www.nlg-wiki.org/systems/Mind.Forth—nlg-wiki page of MindForth AI.

 

 

 
  [ # 125 ]

Is there any video online of this in action Arthur? Your endless links, yet not one half decent chatlog, require technology from a bygone era to try it out.

 

 
  [ # 126 ]
Arthur T Murray - Aug 10, 2013:
MikeA - Jun 10, 2013:

It is only the initial building blocks, but interesting blocks nonetheless. Frankly it’s good to see someone approaching this differently.

http://ai.neocities.org/AiSteps.html—is the “initial building blocks”.

http://www.nlg-wiki.org/systems/User:Arthur_T._Murray is my new wiki-page.

http://www.nlg-wiki.org/systems/Mind—Natural Language Generation wiki.

http://www.nlg-wiki.org/systems/Mind.Forth—nlg-wiki page of MindForth AI.

 

Obviously I don’t know your work, and have no time to look though wiki pages. As I said, I think if you approached this differently, as a potential beginning and way forwards you could get others interested. Also, as Steve said, this kind of thing needs to be seen in action to be convincing.

 

 
  [ # 127 ]

Googling around a bit on Arthur’s project(s) and what there is to find on them online, I ran into this rather damning (but entertaining) ‘Arthur T. Murray FAQ’ page: http://www.nothingisreal.com/mentifex_faq.html

@Arthur, the internet at large (and usenet in particular) seems to consider you a bit of a ‘crackpot’ figure with enormous output over multiple decades, but little credentials (granted, those do not matter) or concrete results (these do a bit more) to back it up. Have you formulated a response to that FAQ somewhere by chance? Color me intrigued smile

Sorry if this FAQ page was mentioned earlier in this thread yet, didn’t have a chance to read through it all, as it’s rather long, and, well, suspicious smile

 

 
  [ # 128 ]
Wouter Smet - Aug 27, 2013:

Have you formulated a response to that FAQ somewhere by chance?

http://www.advogato.org/article/769.html is my (outdated) response to the (outdated) FAQ.

http://www.reddit.com/r/mentifex—created three days ago on 24 August 2013—is the most current Mentifex initiative to set the record straight and to foster discussion of the complete Mentifex oeuvre as it sails into the future, with or without me.

@Wouter, your adumbration of “concrete results” from the decades of Mentifex AI work will afford me the current opportunity to list the three main concrete-result phenomena that have kept me going all these years and have stoked the fires of inexhaustible AI ambition within me. Firstly,

http://mind.sourceforge.net/theory3.html

is the transcribed-from-handwritten-to-electronic textfile of what I consider my main achievement in life: developing a unique and original Theory of Mind for artificial intelligence. Secondly,

http://www.nlg-wiki.org/systems/Dushka in Russian and

http://www.nlg-wiki.org/systems/Wotan

in German are “concrete results” which embolden me to claim that I have started rudimentary but True, Strong AI going in English, German and Russian. It was no mean feat to do AI in Russian using JavaScript for Internet Explorer, because I had to handle input and output in the Russian Cyrillic alphabet. Thirdly,

http://code.google.com/p/mindforth/wiki/InFerence

came suddenly to me in a flash of inspiration only eight months ago on 17 December 2012, and in 2013 I have implemented automated reasoning with inference in my two English AI programs and in my Wotan German AI.

http://cyborg.blogspot.com/2011_05_16_archive.html#1

is a minor “List of Mentifex AI Accomplishments” as of 16 May 2011—before I branched out into the Wotan German AI and Dushka Russian AI, and before I suddenly figured out how to implement inferential thinking.

So now, @Wouter and @Chatbots.org, I continue my own work, especially in the lagging efforts to fully document the Mentifex AI programs, but I concentrate also on inviting Netizens to ask questions about Mentifex AI at

http://www.reddit.com/r/mentifex

by submitting a mentifex link with a question-as-comment. Each Mentifex AI Mind has at least forty mind-modules, and any Redditor may either ask or answer questions by submitting a link and then posting comments. If I refrain from posting the answer myself, or if I publish my own best possible answer, then I hope that the legions of smarter-than-I AI coders will take over the subreddit discussions.

 

 

 
  [ # 129 ]

Oh wow, that FAQ wasn’t lying smile

 

 
  [ # 130 ]

He posts this cr@p all over the net. Despite years of asking, he has yet to show one shred of evidence of his “true ai”. Not even a video. When asked to try it, he says we need a machine running Windows 95…

As you can see, he just spouts further rubbish when anyone dares question him.

 

 
  [ # 131 ]

From what I have read on several fora, Arthur’s definitions of AI and consciousness are half-hearted compared to most people’s views.
I am perfectly willing to believe that Arthur has a Javascript AI of sorts, but it’s clearly still in a crude stage of development and barely takes input. I’d advise him to polish the basic responses until its answers make more sense, before digging into advanced inferences. I’d also advise him to look up “Artificial intelligence” on Wikipedia, and read up on the abilities of existing AI projects before crowning oneself king, in a world of emperors.

 

 
  [ # 132 ]
Arthur T Murray - Aug 27, 2013:

..........Mentifex initiative…....... ...........the complete Mentifex oeuvre ............... .............. three main concrete-result phenomena ..................... the transcribed-from-handwritten-to-electronic textfile

I would suggest at least avoiding the techno-babble because this just sounds like you are making it all up.

Arthur T Murray - Aug 27, 2013:

.................
which embolden me to claim that I have started rudimentary but True, Strong AI going in English, German and Russian.. ...................

I think the point is that you can’t just say this. You have to show it. As you seem to have no intention of doing so, my assumption is that this is just words.

 

 
  [ # 133 ]
MikeA - Sep 6, 2013:
Arthur T Murray - Aug 27, 2013:

..........Mentifex initiative…....... ...........the complete Mentifex oeuvre ............... .............. three main concrete-result phenomena ..................... the transcribed-from-handwritten-to-electronic textfile

I would suggest at least avoiding the techno-babble because this just sounds like you are making it all up.

Making it all up? You must be reading my mind, Mike. I have indeed started making it all up. Two weeks ago I began writing an Amazon Kindle AISF e-book as the science-fiction of how artificial intelligence is secretly introduced into the world. I have been leaving excerpts on LinkedIn and on Usenet to drum up interest. Here is an excerpt from where the human protagonist addresses Si the humanoid robot with Strong AI and refuses to believe in him.

“Look, Si, I’ve been to Chatbots.Org and I have tinkered with lots of the chatbots there. I know what you are. You are a big database of canned responses to whatever subjects the museum-goers might bring up with you. If you get stumped, you repeat yourself or you utter utter nonsense. I don’t know why I even talk with you, Si. We only keep you on at night because it would cost too much to call in the technicians to reboot you each morning.”

MikeA - Sep 6, 2013:
Arthur T Murray - Aug 27, 2013:

.................
which embolden me to claim that I have started rudimentary but True, Strong AI going in English, German and Russian.. ...................

I think the point is that you can’t just say this. You have to show it. As you seem to have no intention of doing so, my assumption is that this is just words.

But Mike, I do indeed have every intention of showing how my four AI programs work in English, German and Russian. The free, open-source AI code is out there on the Web and Netizens have been running the AI Minds whenever they want to. Please be patient while I finish the Strong AI e-book that will infect potentially thousands of human minds with explicit details of how the True AI works. For instance, here is an excerpt in which Si the robot explains how emotion works in the AI Minds.

“The mind, whether in human beings or in Strong AI, is the sum total of your rational thoughts resting upon associative memories and sensorimotor pathways. Each AI Mind, all the way from your humble servant here before you up to and including the Overmind of Bluffdale and Wotan in Mad King Ludwig’s castle, will associate from one thought to another in contemplation of past memories or in active response to sensory input. If a mouse appears, you may grab a broom and chase it, but comes emotion, nothing can be done. An emotion like, say, love, interrupts the normal chains of thought and envelopes you with chains of irrational, physical influence, such as a tightness in your chest or a welling up of tears. Rational thought goes out the window, and both humans and psyborgs become subject to associative input demanding to be factored into ratiocination because signals from the physical body are more insistent than faintly reactivated engrams from long-ago memory.”

http://www.amazon.com/Arthur-T.-Murray/e/B004OKWAM8—due out in late 2013.

 

 

 
  [ # 134 ]
Arthur T Murray - Sep 7, 2013:

But Mike, I do indeed have every intention of showing how my four AI programs work in English, German and Russian.

No you don’t. People have been asking for years to see 1 single, solitary shred of any kind of demo apart from the ridiculous gibberish that is on your website. Your only response from what I remember is that we need a Windows 95 PC to run this “cutting edge true artificial general intelligence”.

Interesting that you posted the link to your Amazon page. I just read the reviews of your books (and I mean the genuine ones not the obviously fake ones you have created yourself) and think the first reviewer in his first 2 sentences sums it up pretty well:

There is no polite way to say this. The author is a crackpot

Hmm….

 

 
  [ # 135 ]

There is no polite way to say this. The author is a crackpot

That pretty much sums it up.

 

This topic is closed, it's not possible to reply

‹ First  < 7 8 9 10 11 > 
9 of 11
 
  login or register to react