When I take a closer look at these dimensions: AHAI, AHAK, AHAB; I begin to see a pattern.  Some people learn by watching.  Some learn by doing. Others learn by conceptualizing.  Would that correlate watching to knowledge, behavior to doing, and intelligence to figuring it out conceptually.  Maybe what we are dealing with here is personality.  And maybe IQ should be defined differently for the kinds of people in the world.  Mozart verses Garcia; Jordan (jock?) verses Gates (nerd?), etc.

Education professionals recognise as many as eight different dimensions of intelligence such as mathematical, musical, movement etc so there is already plenty of formal research being done in that area.

In addition, different people do learn various skills best in different ways. Examples of these are by direct experience, by reading, by listening or by observing. Not sure if this is a function of personality so much as just different brains although there is bound to be a correlation.

Any other ENTJ’s on here?

So Wolfram Alpha, the computation knowledge machine, is intelligent in the dimension of mathematics.  I might be inclined to say it has a refined aptitude for math, as educators would pigeon hole such ranking, but I still don’t feel right about saying its IQ is anything at all.  Yes it can answer the educators’ test questions, but no, it is not “thinking” really.  It is computing, but computing doesn’t equal intelligence. It powers Bing, so that search engine should be intelligent?

From Doug Lenat at CycCorp in 2009: “Stephen Wolfram generously gave me a two-hour demo of Wolfram Alpha last evening, and I was quite positively impressed. As he said, it’s not AI, and not aiming to be, so it shouldn’t be measured by contrasting it with HAL or Cyc but with Google or Yahoo.”

This is the catch between AHAI and AHAK that started the thread. AI is misunderstood. The capability to perform a skill doesn’t make intelligence when it is only a programmed function (like a bad habit.)

Gary Dubuque - Jul 24, 2011:

Yes it can answer the educators’ test questions, but no, it is not “thinking” really.  It is computing, but computing doesn’t equal intelligence.

This may be the crux of the AI problem. No matter what capabilities the computer shows, it will only be “computing”, not “thinking”. It is sort of like how magic is only a trick when you know how its done. Is a spreadsheet or database intelligent in a limited domain?

Perhaps the best we can hope for is:

Is your computer/AI smarter than a fifth grader?

Dave Morton - Jul 23, 2011:

To me, the difference between knowledge and intelligence is the simple premise that knowledge is having data available for recall, but intelligence is the ability to use that data in different ways to solve problems or perform tasks or functions that simply having said data alone cannot accomplish. Anyone can learn that Ohm’s Law is V=I/R, but it takes intelligence to know just what that means, and be able to put that formula to use to build a circuit.

I have tended to partition the “intelligence spectrum” as follows:
data>information>knowledge>wisdom

DATA = Information + noise (all input)
INFORMATION = Components of input that can be used to achieve a goal or make a decision.
KNOWLEDGE = The ability to correctly perform a task based on the information available. Intuition could be considered the ability to perform a task with imperfect or missing information or to create new information through analysis of known information. Creativity might be the ability to create/perform new tasks by combining information in new ways and with imperfect information.
WISDOM = The ability to decide when to correctly perform a task (and when not to).

Merlin, I like your “hierarchy”. What you define as knowledge is similar to what I defined as “creativity.”

Perhaps it might be worthwhile to distinguish the first part of your definition of knowledge,

The ability to correctly perform a task based on the information available.

as a separate (lower) level. Any function performs a task based on input arguments (information available). In fact, this “ability” is certainly required to go from “data” to “information” in the first place.

C R Hunt - Jul 24, 2011:

Merlin, I like your “hierarchy”. What you define as knowledge is similar to what I defined as “creativity.”

Perhaps it might be worthwhile to distinguish the first part of your definition of knowledge,

The ability to correctly perform a task based on the information available.

as a separate (lower) level. Any function performs a task based on input arguments (information available). In fact, this “ability” is certainly required to go from “data” to “information” in the first place.

You may be right. Intuition and creativity may belong on the same level as Wisdom. Each of these deals with parts of the unknown. These are different than knowledge which relates to known elements and processes. I know how to use a screwdriver to turn a screw, but that alone is not “creative”.

Since this was my first thread on this forum I’d like to wrap it up with my observations.

The people who responded are not only knowlegeable and intelligent, but passionate, as well. I appreciate the time and effort you put into your responses.

The subject was an atttempt to determine if we, collectively, have a set of AI terms and definitions in common. It appears that we are still working towards this goal. I’m at a disadvantage. Do to an accident of birth, I don’t speak English. I speak “American”.

“Freedom! This is a sacred word. You will not speak it!” (from Star Trek, the original series). I noticed that some of the human terms we try to apply to computers are considered inappropriate, even by analogy.  A possible solution is to choose new words. Because we are using a natural language, we are allowed to reapply old words, or to coin new terms in an effort to remove objections about the former terms.. For example, we could use the word “cranking”, instead of “thinking”, or coin a new word “bukkolating” to mean the same. Of course, this will introduce a new set of objections. For terms like classifying, interpreting, etc. that aren’t considered intelligence by themselves, I’m considering using the term mental function or process. For knowledge, maybe a broader term, like memories, to indicate data.

I don’t have the answers, just the questions. What I do know is: When it comes to our understanding of AI terminology, the world of AI is not enough.

Well said, Toborman. Well said indeed. I, too, have garnered much benefit from this discussion; perhaps not in terms of advancing my current project, per se, but certainly in the insights that I’ve gleaned from everyone who participated.Thank you all for doing so.

Can I make one more observation?  Before that let me say that I believe a machine someday will be intelligent.  What we have now isn’t intelligent yet.  When machines do, I’ll be entertaining mixed feelings of wonder and fear at the realization that there’s a sentient being in the hardware.

As often it happens, we reach for the low hanging fruit.  Once the easy picking is gone, we don’t find the ladder to climb up and get the rest of the tree’s harvest. And to excuse not making the reach, we call that far away pickings something else.  We claim not to clearly see those rewards because they are too far away. Intelligence would clear up this fog.  Climbing the ladders, we’d see this fruit is not so different than the low hanging stuff.  It doesn’t have to have some fancy new name.  The sad part is that we already know what intelligence is.  We can use the notion quite well in our conversations. It’s only when we decide to “program” it that the confusion creeps in.

What’s that old zen koan?  The centipede was strolling down the road one day when a stranger came by.  Fascinated, the stranger marveled at how all the feet worked in motion to carry the centipede on its way.  The stranger asked, “How do you know which feet goes after which?” Soon the centipede laid struggling in the ditch thinking he did not know.

Can your bot replicate the dialog? That’s the first thing I think of. logicagent:

> Tom is a human.
> Describe Tom.
Default response.

> Describe (.*) is a synonym for what is (.*)
Okay, (?i-mx:^Describe (.*)) has been added, and will now call what_is.

> Describe Tom.
Tom is a human

> Humans are mammals.
> Describe humans.
humans are mammals

> mammals includes a mammal
> humans include a human
> Is Tom a mammal?
Yes, Tom is a mammal.

> why is Tom a mammal?
Tom is a mammal because: tom is a human, and a human = humans, and humans are mammals, and mammals = a mammal

—-

> reset graph
> Tom is a male.
> Tom is a sibling of Gary.
> Tom is happy.
> Describe Tom.
Tom is a male, a sibling of gary, happy

Robert Mitchell - Aug 8, 2011:

Can your bot replicate the dialog? That’s the first thing I think of. logicagent:

> Tom is a human.
> Describe Tom.
Default response.

> Describe (.*) is a synonym for what is (.*)
Okay, (?i-mx:^Describe (.*)) has been added, and will now call what_is.

> Describe Tom.
Tom is a human

> Humans are mammals.
> Describe humans.
humans are mammals

> mammals includes a mammal
> humans include a human
> Is Tom a mammal?
Yes, Tom is a mammal.

> why is Tom a mammal?
Tom is a mammal because: tom is a human, and a human = humans, and humans are mammals, and mammals = a mammal

—-

> reset graph
> Tom is a male.
> Tom is a sibling of Gary.
> Tom is happy.
> Describe Tom.
Tom is a male, a sibling of gary, happy

I like it.  Thanks for the reference to “Logic Made Easy”, by Deborah J. Bennett (c. 2004).

Toborman - Jul 20, 2011:

In a recent forum someone stated that AI developers had not reached their AI goals due to the lack of a large database of knowledge. The response was that a large database of knowledge was not necessary to accomplish AI goals. This led me to observe that when one does not distinguish knowledge from intelligence then one cannot determine the value of knowledge to intelligence…

To simply answer your wish to distinguish knowledge from intelligence: Intelligence creates knowledge.  Knowledge does not make intelligence.  Knowledge is an artifact of intelligence which is why we can use it to test for IQ.  There may be other indicators for IQs, like the correlation of response time to optical stimulus, but traditionally we have used solving problems.

Is Yago2 not a large db of ‘knowledge’ as you say? There is also framenet and verbnet which describe data in interesting ways.

Finally, isn’t a library just a large information database, stored on paper?

I look at a library as more of a collection of databases, with each book and/or periodical being it’s own “stand-alone” database. Of course, I’m probably a bit off base there, but…

Gary, I like the way you distinguish knowledge and intelligence. Intelligence is the ability to recognize, structure, and manipulate pieces of information. You need intelligence to store/generate/use knowledge, so testing knowledge is a simple way to tell if a human is smart. Unfortunately, computers are great at storing knowledge, so one needs to be more careful about how to test for intelligence. I think that’s why the first thing many people do when they talk to a chatbot is try word problems and the like to test the program’s ability to manipulate new information. (“Create knowledge” as Gary said.)

Jan, I hadn’t seen YAGO2 before. Thanks for the link!