Hello Juan. I think that we need to clarify what to understand by "artificial". When I consider a system that performs certain functions, I distinguish between the sources of the ability to perform. If the system was endowed with the ability by man, to me, it is artificial, otherwise natural. Given that scientists have produced automata and we live in the age of robotics, it must be noted that every stage in the development of such systems calls for the intelligence of an agent. I recall that in the early 1970s, the computer we had at my college of technology needed many lines of algorithm to draw a straight line, and four such inputs to draw a square. Given that a computer (or any robot) cannot do anything that it was not programmed to do, I cannot think of artificial intelligence without an agent. If you extend this concept to everything that exists, objects, animals, or humans, one cannot sensibly deny the existence of an intelligent designer..

Dear Juan-Esteban Palomar Tarancon

Regardless of what they are called – agents, intelligence, machine learning, AI - despite my tongue in cheek statement, beating CAPTCHA and CAPTCand learning to calculate sum of two number or to play video games are significant technical achievements that are foundational to many powerful applications that will change the multiple industries.

Kind Regards

Qamar Ul Islam

Dear Jaun,

Basically what we are trying to achieve today and what intelligence really is, are very far apart. Whereas the machine performs tasks on the basis of a goal-based approach to measure intelligence. It is hard to attain true intelligence through machines.

Take into account that my question is semantic: What we have to term artificial the device or its operations?

The operations do perform by the devices are based on the underlying algorithms. It is an algorithm that enables the devices to perform the operations intelligently.

I would like to define the terms first to make my answer more lucid.

Artificial Intelligence is intelligence demonstrated by machines, unlike the natural intelligence displayed by humans and animals, which involves consciousness and emotionality. AI refers to the simulation of human intelligence in machines that are programmed to think like humans and mimic their actions.

While an AI agent is an autonomous entity (software/software embedded in the robotic hardware) that senses the environment (surrounding) and take an action using actuators on the environment to achieve a well defined goal. Nowadays the agents learn, take action, get a feedback and then take the next action. Prominently known as reinforcement learning.

When we speak of a Computer - it's a machine that can be programmed to carry out sequences of arithmetic or logical operations automatically on executing the variable program.

All agents are programs, but not all programs are agents.

Four key notions that distinguish agents from arbitrary programs: reaction to the environment, autonomy, goal-orientation and persistence.

Thus, the computer solving an arithmetic operation on receiving an input and generating an output sum via the machine level codes and logic gates in the ALU, PR, AD components in the Central Processing Unit is a mere system not an agent.

As the computer programs are less autonomous, lack social behaviours and are less reactive or responsive and do not have a specific goal.

Even a calculator is a simple machine but it is not an autonomous agent or a bot.

Thankyou for the question.

Dear Raksha,

You say "Artificial Intelligence is intelligence demonstrated by machines, unlike the natural intelligence displayed by humans and animals,".

Suppose that a machine calculates 2+2=4, and a human also calculates 2+2=4. Is the first expresion 2+2=4 artificial and the second one 2+2=4 natural, or artificial is only the machine?

You are right when say "Artificial Intelligence is intelligence demonstrated by machines," my question is whether this is a proper definition.

Hi Juan. Now I think I have come to understand your question better. You are looking at the same output from different sources, man vs machine. It is quite unlike producing an artificial flower instead of a natural one. In your case, the output is the same and come under the same qualification. There is nothing natural or artificial about mathematical expressions. They are conventions to help us quantify phenomena regardless of the source. It has been very clearly pointed out that "intelligence" refers to the source of an activity not to the activity or output. Thus, 2+2=4 in itself Is not intelligence, but the ability to work it out.

Hi Juan, I agree with Akpan Jimmy Essien answer concerning the mathematical expression. Also, regarding the definition of :

Artificial- imitating not naturally occurring, not human-generated.

Intelligence - a source's ability to exhibit human mind, cognitive and social behaviour.

Now we have 3 entities -> 2+ 2= 4 :

Humans--> Nature created minds ability to solve a sum on training.

Computer programs, not bots -> Human created an imitation of less autonomous cognitive behaviour

Agents(bots, humanoids)- Human created an imitation of more autonomous cognitive behaviour with additional social and learning capabilities more proximal to the mind.

Hi Raksha,

Would you write a constructive or structural definition of intelligence?

To say that intelligence is what a machine or a human does is not a constructive or structural definition. For instance, if you say that houses are what architects make, and I see some architect cooking a cake, I can interpret that cake = house.

It is AI agents that galore!!!

They have the basis of Natural Intelligence.

The computer is human-made. That makes it artificial, since it’s an artifact. The programming is also handled by humans, meaning that is also human-made as in artificial. What the computer programming produces is merely the result, since it wasn’t created by us, but by the computer. The means to produce it, both the physical computer and the programming, are artificial, but not the results.

From the dictionary:

artificial | ˌɑrdəˈfɪʃəl | adjective 1 made or produced by human beings rather than occurring naturally, especially as a copy of something natural: her skin glowed in the artificial light | an artificial limb | artificial flowers. • (of a situation or concept) not existing naturally; contrived or false: the artificial division of people into age groups. 2 (of a person or their behavior) insincere or affected: an artificial smile. 3 Bridge (of a bid) conventional as opposed to natural. ORIGIN late Middle English: from Old French artificiel or Latin artificialis, from artificium ‘handicraft’ (see artifice).

I recommend reading "Dialogues of Plato". When intuition substitutes logic, we get into a discourse that leads to nowhere. Everybody believes to know what intelligence is. Without stating a constructive or axiomatic definition of intelligence, this discussion leads to nowhere.

I very clearly referenced the dictionary to introduce the formal logical definition of “artificial.” As such I did not act on intuition, but stuck to sound logical and scientific principles in answering your question.

On top of that in order to stick to a “dialogue,” you would have to answer to my answer, rather than avoiding any kind of answer to the content of my writing. I sincerely doubt Plato would teach evading a proper dialectical discussion of thesis, anti-thesis, and synthesis.

My answer was clear and concise and in no way didn’t lead anywhere: it lead to a very straightforward conclusion without meandering on whimsical thought.

In terms of intelligence, if that’s the only thing you wish to discuss, a computer has no understanding and always requires a human to interpret or build something for it to interpret its answer. Even when a machine has interpreted the answer for us by printing it for instance, we are still the ones that have to make use of it somehow of which the computer can never have any kind of knowing.

I know what is said artificial or natural. My question is whether this definition fits into the scientific method. I think that artificial is the device and not the product.

See a simile. A kg. of honey K is produced by a machine and another one H by a honeycomb. If the chemical composition, flavour, and attributes are the same in both samples, the scientific method identifies H with K. The first method is artificial and the second one natural, but the honey is the same. When a human and a machine apply an algorithm, the algorithm is the same. What is not the same, is the intelligent system. This is why I think that the proper denomination should be AIS = Artificial intelligent system, instead of AI.

That’s also the answer I gave, that artificial equals artifact or device.

In case of the honey I should point out we didn’t make the bees and the bees have intent of their own. The bees make honey and all animals, including themselves, eat it. It’s a product of their own, they officially only intend for themselves.

The machine doesn’t regard what the honey is for. When a machine makes honey, it’s up to us to decide what its purpose is.

Science should also recognize intent, the dynamics of the product if you will, and not just what is, which is static. The machine made honey has no purpose of its own, even when it has the same chemical constituents. The honey made by bees does have purpose.

If we define objects by circumstances, I can state that 2 ≠ 2, because my wife has written the "2" at the left-hand, but I have written the "2" at the right-hand.

I didn’t say circumstance, I said intent. Circumstance isn’t a synonym of intent.

Sorry,

Intent is not asynonym of circumstance. I agree. What you have write is:

"In case of the honey I should point out we didn’t make the bees and the bees have intent of their own. "

I cannot say anything. I have never being a bee and canot know what a bee thinks.

Sorry for typos: Being = been.

We didn’t make the bees, therefore they aren’t artificial. The bees have evolved to produce honey that they then feed to their larvae, which is the intent of the honey. They have shown that they communicate and form a collective consciousness when necessary to supply and defend the hive.

Is it so difficult to understand that my question is semantic? If in some subject the term "artificial" means red painted, you must understand artificial=red in its scope.

What I have said is that the ordinary use of "artificial intelligence" denotation, perhaps does not fit into scientific method. What is artificial is the intelligent system, not its procedures are. I am able to imitate the working way of a machine and my mind does not become machine.

Our eyes, our hands, our legs and so on are driven by algorithms stored in our brains. When our brains handle these algorithms, does not become a machine. It keeps being the same organ.

Is it so difficult to understand that you need to reference my very first answer to see that I focus very specifically on the semantics by actually quoting the dictionary to create common ground in terms of definitions? As you will see -- at the end of my reply I once again include a dictionary definition, in this case of the word “semantic” -- it’s not me that breaks with proper semantics.

My problem is that

(1) either your conception of language is thus bad that you are not able to engage in the proper interpretation of what’s being said nor to formulate concisely what it is you mean to say, meaning that you haven’t properly semantically formalized what it is you mean to say, or

(2) you simply seem to take delight in sabotaging proper discussions by evading the proper interpretation of what is said to instead introduce notions that in no way relate to what I said or you mean to say, sometimes even saying that what I just said is what you mean to say as if I didn’t say it.

I can’t tell whether you undermine the very fabric of science through your ignorance, lack of understanding, or by your own volition. I can’t see whether you managed to convince yourself that you can do this job. I am telling you, I don’t find you fit for the field of science and if you count yourself a scientist, it seems you don’t even know how to express yourself and do see yourself in ways you are not.

semantic | səˈmæn(t)ɪk | adjective relating to meaning in language or logic. DERIVATIVES semantically | səˈmæn(t)ək(ə)li | adverb ORIGIN mid 17th century: from French sémantique, from Greek sēmantikos ‘significant’, from sēmainein ‘signify’, from sēma ‘sign’.

"Semantic, adjective relating to meaning" ...O.k. Can you write a definition of the term intelligence, before adding the adjective artificial?

intelligence | ɪnˈtɛlədʒəns | noun 1 the ability to acquire and apply knowledge and skills: an eminent man of great intelligence | they underestimated her intelligence. • a person or being with the ability to acquire and apply knowledge: extraterrestrial intelligences. 2 the collection of information of military or political value: the chief of military intelligence | [as modifier] : the intelligence department. • people employed in the collection of military or political information: French intelligence has been able to secure numerous local informers. • military or political information: the gathering of intelligence. • archaic information in general; news. ORIGIN late Middle English: via Old French from Latin intelligentia, from intelligere ‘understand’ (see intelligent).

I appreciate your answer. Without knowing what meaning you assign to the term intelligence, it is not possible any discussion.

When I say that some definition does not fit into the scientific method is by the following reason.

Once we have a definition for the meaning of any word, to apply it to some object, we need to know whether the attributes of this object satisfy the definition. Only algebraic objects can be defined without any ambiguity.

Many atoms, like carbon, oxygen and so on, have la capability of associate forming neurons, and then can acquire knowledge and skills. I think the these atoms possess intelligence: natural intelligence.

Now I need to ask you to define intelligence.

Hi Juan. I think that this discussion is assuming qualities that will preclude distinction between what is "natural" from "artificial". It is rather hard to understand how you came to the supposition that atoms can develop neurons and acquire knowledge. It seems that this fallacy of evolutionary biology is trying to bend this discussion to the point of according anthropomorphic capacities to purely chemical components. If atoms can develop neurons and acquire knowledge, we might also conclude that all artificial things will turn into natural in due course. You've lost me.

Seems to me like Juan already has an answer to his questions. My question for him is what he means by “agent” in his questions. Agency can be attributed in so many different ways; to man, the machine, and some would even say God. How far does one advance or extend agency? Can an operation be an agent? Perhaps tackling the agency issue might shed some light on the issues others have attempted aptly so far.

There are terms, like intelligence or God, that everybody assigns meanings that need not be the same. In science, we identify objects by their properties or attributes. I do not know any constructive or structural definition of intelligence. Every definition consists of describing its effects or actions. If we define X for being the cause of my father's dead, X can be a virus, a killer and so on. Such a definition defines nothing.

The aim this discussion is the distinction between artificial or natural. I assume your non-constructive definition of intelligence.

Suppose that we construct a machine M with the same capability of acquiring knowledge and the same skills that W. A. Mozart. Do you consider that the intelligence of M is artificial? You are free of it; however, the intelligence of M and the one of Mozart are the same. They have the same skills and properties. By scientific method it is not possible to state a distinction. What is artificial is M and not its intelligence because we agree that both M and Mozart have the same skills and capabilities.

When, under logic laws, we say A = B, we mean that swapping A and B it is not possible to see any difference. If we are ruled by logic laws, when the capabilities of A and B are the same, both have the same intelligence. What could be different is the physical structures of A and B, their origins or the way that they are built. This is why I consider that "artificial intelligent system" could be the proper denotation.

Hi Juan,

I would think that when we say, "artificial intelligence," the attribute "artificial" is known to apply to the performing "agent" or "system" and not to the product. So, we have an "artificial" agent which contrasts with "human" or "natural" agent.

Both are performing "systems"---the one natural, the other artificial. Therefore, if you add "system" to "Artificial Intelligence," you must equally add "system" to "Human Intelligence." Thus, the addition of "system" to AI would not change much, if anything at all.

Ji, Akpan,

"Therefore, if you add "system" to "Artificial Intelligence," you must equally add "system" to "Human Intelligence." Thus, the addition of "system" to AI would not change much, if anything at all."

Of course, I agree. Let me tell a joke.

When you say A = B, you really mean to say that if you have agents Agent1 and Agent2, say Mozart and a robot, the attributes for a particular task which you label intelligence equalling ability to perform a certain task mentally, may be equal, but they are only attributes of a larger being.

The totality of one being should be described in terms of a three dimensional construct: physical, social, and psychological. As such:

Mozart->psychology.musical_ability = robot->psychology.musical_ability

AND

Mozart->physical.musical_ability = robot->physical.musical_ability

Although some abilities may be the same, the full being isn’t. Not even psychologically. Although it has acquired the musical skills of Mozart, it doesn’t have the full psychological construct of his being.

Robots don’t have qualitative experience, so although you might train a neural network to come up with proper compositions, it won’t be able to tell whether an instrument sounds nice, as in right for the part. Instruments have undertones and overtones, and Mozart wouldn’t leave room for interpretation. He wrote musical pieces for particular instruments also, for as far as I know, taking the sound of those instruments into consideration.

I’m a little lost since you now introduce arguments that were first brought to you but by the discussion as you engaged in it, you wanted to refute them and said that that wasn’t what it’s about.

Hi Juan. You said:

Of course, I agree. Let me tell a joke.

That is very interesting. The joke itself is self-contradicting. If God has best distributed intelligence to everyone according to you, why would everyone not be right to think that he has enough? Besides, how does this relate to your question on artificial intelligence systems?

As A. Imam has noted above, "Juan already has an answer to his questions." Thus, for me this matter is resolved. I'm out.

Hi Akpan,

The sentence is not a theological or philosophical dogma. It is a joke that it is not easy to understand without having Spanish sense of humor. What the joke means is that stupid people feel themselves very intelligent-