Terminator Genesis 7 – A World No Viruses!

On one beautiful day, I received an email notifying me of a new Facebook post by my best e-friend, Bishop Dr. Ijaz I. Malik.

He warned people and his loved ones from a hideous virus that had hit his computer, deleting his precious video clips.

Doesn’t this seem rather a familiar story?

All the world’s antivirus manufacturers have recently made a disclaimer notifying their clients of their inability to guarantee against any damages caused by viruses.

See for example reports by  ZDNET,  Wall Street Journal,  MIT,  CNET  and  The Guardian

In fact, we do not need so many proofs because currently, the computer will never see “A VIRUS” as “A VIRU5” i.e. The slightest jot can result in a disaster.

Please look closely: the second word’s last character is the number 5, not the letter S.

The question is: Till when shall the computer remain a toy in the hands of the enemies?

In other words, when shall the computer eat from the Tree of Knowledge of Good (useful) and Evil (harmful)?

If your computer gains wisdom and comes to know the good from evil, he will serve you without instruction or supervision.

And on the other hand, he will not cheat you or become a tool in the hands of the enemy.

Now there have been many projects of AI such as OCR (Optical Character Recognition), Speech Recognition, Expert Systems etc.

The latest is ANN (Artificial Neural Nets).

The world is investing billions of dollars in ANNs and big business is making many times more on these investments.

But do these intelligence products understand what they’re doing? Do they got the knowledge of good and evil as so clearly understood by the Terminators?

One theory says wisdom is not quantum. Hence, binary computers will never become wise.

Another theory says wisdom is organic: we can’t create wisdom outside neural nets.

Yet another theory says wisdom is biological. It can’t exist outside a living tissue.

There must even be someone out there who believes wisdom is spiritual. It’s not of this world but a God-given gift — hence, man will never create it.

To answer these questions and give the world some hope, we need to create a PROTOTYPE.

The Prototype may not do everything that Terminator T-800 Model 101 – our subject of interest- could do.

But if it did just a couple things, it will serve as PROOF OF CONCEPT.

It will put an end to endless debates over the nature of intelligence and the possibility of (some day,) creating the Terminator.

Moon Laboratories’ E2P2 (Electronic Ear Prototype Project) will serve an important academic purpose. The Prototype, therefore, will be about research in BASIC SCIENCE.

But that’s not all.

The E2P2 will also show the world that AI is not the exclusive domain of Big Business and a threat to human labor (unemployment & job eradication).

Shortly, I’ll create a crowd-funding project using websites like KickStarter.com and Indiegogo.com.

The purpose of that fund-raising project will be to first create the E2P2 prototype and if all goes well, to also start building the Terminator — one small piece at a time.

I’ll work part-time on Moon Labs’ E2P2 project as I also got other projects that are equally important both for me and for mankind in general.

By participating in funding my revolutionary approach to AI, you’ll be entitled to complete copy(s) of all the building blocks of Terminator at discounted rates.

Those who’ll not participate in funding E2P2, will not get complete copy or will have to pay a premium.

Have a nice day.

Advertisements
Aside

Terminator Genesis 6 – A Criteria of Intelligence or of Wisdom?

Suppose you write a piece of software that can add any given pair of numbers.

Will that software be considered intelligent or dumb?

Can a dumb object like a chair add two numbers?

Isn’t the ability to add numbers a valuable skill? Isn’t addition a mental ability?

So why your piece of software is not considered intelligent by the AI scientists?

What does “intelligence” mean?

Let me surprise you.

The white man actually lacks wisdom. By wisdom, I mean the ability to understand what a chair is. What a fan is i.e. The ability to see things and recognize them.

That’s why they have struggled since the times of Aristotle and Socrates to understand what wisdom is. They decided to call it “common sense”.

Not that this 3000 years of illustrious philosophical venture has helped them gain any wisdom.

They’re still struggling to understand what “common sense” means.

Just because your piece of software got a lot of intelligence, it doesn’t mean it’s “intelligent” — because it got intelligence but no wisdom:

. It doesn’t know what numbers are.
. It doesn’t know what “addition” is.
. It doesn’t know what’s the use of adding numbers.
. It even doesn’t know whether it should add those numbers or, say, multiply them.

Heck. It doesn’t even know it’s adding numbers. It can’t choose not to add the numbers. It can’t choose to err and give you a misleading answer!

It got no mind of its own. Your piece of software is not WISE.

And wisdom is defined herewith as “The ability to receive sensory data, to correctly perceive it, to discern its implications and to make (independent) decisions that maximize a profit”.

Since your piece of software adds quickly and accurately, it’s definitely intelligent. In fact, it’s more intelligent than any human being but it’s not wise — because it lacks any of the abilities mentioned in The Definition of Wisdom.

It can’t even compete with a single-cell creature. Because even a single-cell microbe knows hunger and it tries its best to find food, for example. A microbe is wise.

A computer may beat humans at Go (the famous board game) and be called “A major breakthrough in AI” but does it know what it’s doing? Does it like playing Go? Does it know why to win? Can it choose not to win?

What kind of breakthrough people are talking about?

The scientists have convinced the investors that if we continue building larger & more powerful artificial neural nets, we’ll somehow achieve Strong AI (Wisdom).

They believe we’ll just wake up one day and the computers will be wise. They’ll listen and understand. They’ll look and see. Stephen Hawkins even warns us that artificial neural nets may one day learn stuff we never intended it to and turn against us!

The truth is bitter.

The truth is that no matter how advanced an artificial neural net is, the computer will still remain a dumb machine executing (machine) instructions without having the least idea of what it’s doing. Or the least interest in doing it!

I suggest either using “wisdom” as defined above or coming up with another word that fits The Definition of Wisdom.

I’m a Pakistani and since they got wisdom, they have words to describe what computers lack.

They know what makes computers do dumb things. Like catching a virus and doing damage to themselves and others.

The kind of wisdom I’ve defined her is called “Akal” or “Samajh” in Urdu. Computers lack akal — all Pakistanis know that.

It’s one reason why I succeeded in developing the necessary algorithms to make a computer understand what it’s looking at, what it’s hearing, to whom it’s talking, what’s going on, whether doing something is profitable or not and therefore, decide whether it should do it or not.

However great my achievement may be, it still needs to be converted into machine code. And I lack necessary resources.

I’ve already requested contributions to help me code and feed my algorithms to computers. Please spread the word to your network of family, friends and business associates.

I’ll also try to raise funds through Kickstarter.com.

If you read my other posts, you’ll know that my rich brothers are of no use — they want me dead because I dared to marry against their choice.

Aside

Terminator Genesis 5 – The Role of Definitions In Solving Problems

In the previous post titled  Terminator Genesis 4 – The Importance of Definitions,  I explained how all natural creatures depend on their ability to construct definitions to see things.

And when I say “see”, I also mean to hear, taste, smell, touch or see abstract things such as love, intentions and work.

In this post, I’ll show you how definitions also help in understanding problems as well as in solving them.

I’m not going to show you HOW to build an electronic problem solver. I’l just show you the important role definitions play in solving problems — just as I only showed you the importance of definitions in the work of our senses but I did not show you HOW one can build an electronic definition builder and hence, electronic sense(s).

If you are good in mathematics, you should already know the fact that each problem (and solution) is a mere definition but I’ll assume that you’re not a mathematician.

Thus, you’ll be surprised to learn that a computer program is nothing but a definition and that a nest is also a mere definition.

A sparrow can build and maintain a good nest because it got a very good definition of what a nest is.

What it does to build a nest is to work towards constructing a nest that fulfils the definition it has in its mind, using its skills and its faculties. (I’m not going to discuss these skills and faculties yet.)

If you remove a part of its nest, the sparrow will know because the nest will no longer fit the Definition. It, therefore, will start repairing the nest until it again fulfils the Definition (assuming it decides to.)

Again, if you throw some paper balls in its nest, it will remove them to keep its nest clean & tidy.

In the first instance, we made the nest incomplete and the sparrow realized that and repaired the damage to re-complete the nest.

In the second instance, we added stuff to the nest that’s not part of its definition. Again, the sparrow realized the problem and returned the nest to its proper state.

A software developer works in a similar fashion.

Suppose that you asked a programmer to write a code to print the numbers 1 to 10 on the screen.

Then the programmer will start constructing a program that fulfils the problem definition.

At this stage, she’s like a sparrow constructing its nest.

After writing the code, if you delete some part of her program code, she’ll know it and will replace that code.

Similarly, if you add something wrong to her program, she’ll know it and remove it because she knows what the correct program should be.

Isn’t it beautiful how computer coding is like building a nest?

If we assume that all problems’ solutions in the world are like coding a computer, then by building a definition builder, we’d not only have built an electronic ear, but also have built a general problem solver!

A true general problem solver.

Let me inform you that the algorithms used to build an electronic ear will also be used to build the electronic eye i.e. our algorithms will be a general sensory data (signal) processor.

I got the algorithms for analyzing an audio (or video signal) and creating the definitions. These algorithms can also refine the definitions with experience.

To convert these algorithms into computer code, however, I need an investor who’ll finanace my work for at least two years.

An electronic ear can be a good addition to any computer device but it’s particularly useful for small computing devices such as smartphones and tablet PCs.

According to  Statista.com,   there were 1.86 billion smartphones in the world in 2015, projected to reach 2.87 billion by 2020.

The  Pew Research Center  estimates 68% of Americans owned a smartphone and 45% owned a tablet PC in 2015.

Hence the potential market for an electronic ear is quite large.

Thus, my financial supporter will be given handsom returns for sharing the risk(s) with me.

A typical investor will pay $339 upfront + $79/month thereafter or $479 paid half-yearly. I expect producing the first version in 2 years.

In case of success, the typical investor will receive twice their money. The larger your investment is, higher the return you’ll get. . I’m open to suggestions.

I’ll next further discuss definitions. Till then…

See Ya.

Terminator Genesis 5 – The Role of Definitions In Solving Problems

Terminator Genesis 4 – The Importance of Definitions

If you try to hit a mosquito with your hand, it will fly away.

This means that even a mosquito knows what a hand is, what damage it can do (crush), that getting crushed is a life hazard and that life is precious that doesn’t return once lost (and is worth saving).

A mosquito understands what your hand is, what getting crushed is, what a hazard is, what life is, what loss (of life) is.

In other words, a mosquito got the correct definitions of all above.

Without having correct definition(s), one cannot correctly identify or understand what something is.

Let me give you an example.

Suppose you’ve never heard about apples and that I’m trying to teach you to recognize an apple when you see one.

If I gave you, say, this defition of an apple: “an apple is something round and red.” Then you’ll also consider a red ball an apple — which is wrong.

Suppose that I then improve upon my definition and say “an apple is a round & red object that is edible”.

Then you’ll reject a yellow apple and consider a tomato an apple — both of which are mistakes.

Hence, my definition is not yet concise & correct.

Suppose I then say “an apple is a fruit that is usually round and tastes like apple”, then this definition will not only exclude all objects that look like an apple but do not taste like an apple, but it will also correctly identify, say, a genetically-engineered black apple.

Even if you eat a dish or drink a beverage made of apple, you’ll still know that it contains apple.

Let me clarify that “tastes like apple” does not constitute a recursive (circular) definition because the apple’s flavor is not an apple but it’s a chemical that can be completely defined by a chemist independent of apples.

Assuming that the last definition was perfect, you’ll now correctly identify an apple when you see one, even when it’s altered.

The mosquito had correct definitions when it decided to escape your attempt to crush it.

Now let us see whether a computer can “see” anything.

Suppose that we use a camcorder to record a session between a patient and a physician.

Now if a human is shown that recording, she’ll correctly understand what’s going on. She will also identify doctor’s clinic, the doctor, the patient, any objects on doctor’s table, the chairs, the walls and any other object known by common people.

So it’s safe to assume that there was nothing wrong with the recording.

But what if we show this recording to a computer? Will the computer be able to understand anything? Will it be able to identify any objects?

No. The computer can record the video and play it back but it will not understand anything — and it’s terribly difficult to teach it to identify simplest objects.

And it’s identification performance will always remain awful. Compare that to a tiny mosquito’s!

Another serious issue with the computers is that they can’t learn about new objects.

Many domestic animals can identify common household objects and they can usually also understand many abstract “objects” such as “death”, “food”, “injury”, “love” and even “work”.

If you’re, say, a software developer and got a cat, then the cat will initially not understand why you’re wasting your time on a computer — it’s not food.

But after a while, the cat will understand that your computer is actually a tree whose fruit (money) you use to provide for your food (as well as cat’s).

Conclusions:

1. The ability to “see” things is dependent on our ability to construct definitions.

2. To achieve AI (Artificial Intelligence,) we need definitions, not a neural net to compare (a stored image) to another (that we expect the computer to be able to identify).

3. Even a mosquito with less than a milligram of gray matter knows what is “food” and thus, can find it.

It’s not because its brain is more powerful than a supercomputer (or even an ordinary smartphone’s processor). It’s because it uses different algorithms.

4. A computer equipped with a camcorder is capable of vision and hearing — provided we let it eat from the “tree” (algorithms) of vision & hearing.

5. The computer lacks no intelligence. Intelligence is the number of questions one can solve in a given time. An ordinary person can solve 100 questions/hour. A computer can easily solve millions and billions.

6. We first need to solve the problem of computer vision and hearing. In other words, we first need to build an electronic eye and an electronic ear.

And for that, we’ll also need to build an electronic brain and a database so that the computer will then learn to see and hear.

Aside

Terminator Genesis – 3

Now that we have a computer that has enough sense to keep “awake” as much as possible, to immediately sense new devices as soon as they’re attached to it and immediately activate them, to start crying louder & louder as soon as it loses power supply and also to keep a record of all its “experiences”, we now need to teach it how to understand that input and learn from those experiences.

At this point, the computer will be said to have acquired FIRST ORDER INTELLIGENCE. This is for future reference.

We’ll need to assume that the Experience Recorder got compression capabilities and our machine’s storage is unlimited.

Not everyone can afford to build such a machine!

Now we need to re-define “Artificial Intelligence”.

I’ve already presented an apology for our need to re-define AI. “Artificial intelligence” is a misnomer and has misled scientists as well as investors because computers obviously lack no intelligence, do they?

What a computer can do in a matter of seconds, we humans can’t do in our life time. For example, Google Search can mine billions of webpages in milliseconds and find that one page which contains the search term “moonlabs terminator”, for example.

Isn’t that impressive? Who can contend with that giant?

Or you can use an Excel sheet and fill a column with 1000 random numbers and then use the SUM function or even multiply all those 1000 numbers together and any computer will compute the results in seconds.

Who can match such a speed and accuracy?

We, humans, cannot match any computer (or even a pocket calculator) neither in speed nor in accuracy.

So what do computers lack that made my brother complain to me?

. Computers lack understanding & common sense.

. They don’t have common knowledge.

. They are inconsiderate.

. They can read input text but they don’t understand it.

. AND THEY NEVER LEARN. They need to be programmed and re-programmed every time — a painful & costly job.

They are giants who’ll obey all your commands but they got no mind of their own.

And when I say “no mind”, I mean totally mindless as in “zero brains”.

So if you enter the command 7*5 but by mistake type 7 x 5, then the computer will output some curse words instead of 35.

And you can’t blame it.

So when a computer misbehaves, it’s always someone else’s fault! Sounds like my wife!

But how can we teach computers some understanding?

Scientists are divided on whom to call the Father of Modern Mathematic. I tell my students Archimedes was the good father of modern mathematics because he said “Mathematics is a science from heaven”.

That’s a great insight.

“Mathematics” is a Greek word that means “to learn” so God has given us the ability to learn and put us in a world that is the largest school & longest movie script running continuously since some 6000 years.

The local Greeks had translated my Father’s kingdom, “Eden”, into “school”. They were wise and made a true translation. It later transmuted into current Skopje but that’s another story.

The other “gentiles” were not as wise so they say “What’s in a name?” I don’t know why they don’t care.

I am a mathematician and a philosopher and I love wisdom. I don’t like misleading names like Artificial Intelligence. But I’m still going to use it because it has stuck since decades and there is no alternative in English that will convey what we’re trying to develop. “Artificial Cognition” comes close but is not good enough so let’s continue the traditional name.

Dumb computers are quite fine.

We find dealing with image and sound the easiest, dealing with words harder and dealing with numbers the hardest so we built machines that are number crunchers.

Hence, writing programs that deal with numbers is the easiest. Programming for words & letters is a little harder and programming for images and sound is the hardest.

We and computers are opposites (and complimentary). It’s good to have a villain!

But now we need to build a computer that can understand images and sound like us.

For that, we need to teach computers the art of mathematics (building definitions).

If a computer can construct definitions, then it can understand.

I’ll show you examples to prove this crucial assertion in the next post. Till then…

See ya.

Terminator Genesis – 3

Terminator Genesis – 2

In order for a computer to behave like a human (or any natural creature for that matter), it must not be passive, waiting for a user input.

Here is what a typical computer looks like:

1. You plug it in.
2. You turn it on.
3. It loads an operating system.
4. It stops, waiting for your input. Like this:

C>

Or this:

$_

All that Hamid sees is an intimidating, cryptic system prompt.

In modern computers, you’ll usually see a cold wallpaper called “desktop” and some sort of a “start” button.

You may also hear a musical tone.

This is pretty dumb.

So first thing that we need is to teach the computer to remain “alive”.

The second thing that makes computer dumb and intimidating is their nature of waiting for user action. Even if you plug in a microphone, it may remain passive.

A computer must keep an eye on all its assets, especially any sensors like a microphone. As soon as a microphone is plugged in, it must automatically install its device driver AND turn it on.

It is very dumb to have a microphone yet remain deaf, isn’t it?

The same holds true for web cams, scanners, keyboard, mouse etc.

Now a computer is also considered dumb because it keeps forgetting stuff. It feels as if it has no memory.

What we need next is an “experience recorder”.

We need the computer to record all data from all input devices.

This Experience Recorder must allow access to the streaming data as well as all previous record to other algorithms that’ll process the input data.

It’s also very dumb on the part of computers to use energ-saving habits and go to sleep or hibernation mode without regard of others.

Our machine must remain turned on at least during day time.

In fact, it must have as large a UPS as possible and as soon as it’s plugged off, it must start crying for “food”.

Now with these features in place, the computer won’t be as dumb as the ones my brother, Hamid, hates.

The computer is still not aware nor able to understand a thing. For that, we’ll need to study how natural creatures gain awareness and develop algorithms to mimic these skills.

Till next post,

See Ya.

Terminator Genesis – 2

Terminator Genesis 1 – The Need For Terminators

Once upon a time, there were Right Brothers. They built a machine that could fly and the machine had little resemblance to how birds fly.

They were not alone.

The printing press before them had little resemblance with our hands, the natural writing tool.

Still, there’s a particular aversion among computer scientists against Natural Intelligence.

I find this aversion peculiar.

On the one hand, these scientists are obsessed with neural nets because of the neurons that make up natural brains. And on the other hand, they dislike any discussion of NI or attempt to mimic it.

I know they do this because they don’t understand NI and they’re worried about the flight of capital from their field of preference.

Now let us revisit the first problem faced by Hamid. He wanted the computer to print the value of Force, given Mass and Acceleration.

How hard that could be?

Here is the FORTRAN pseudcode he wrote:

Force = Mass * Acceleration
Mass = 5.0
Acceleration = 3.0
WRITE (Force)

He was annoyed to see the computer’s inability to calculate Force. Instead of printing 15.0, it kept printing garbage.

The reason is that computers do not have a memory like ours. They are especially designed to process data in a sequential manner and once they’re done, to forget everything.

So when the computer looks at the above code, it will multiply Mass by Acceleration and store the result in Force.

And at this time, it won’t look at nor see the next two lines where the values of Mass and Acceleration are given. Hence, according to the FORTRAN specifications, it will multiply any garbage value in the memory location assigned to Mass with another garbage in Acceleration, producing yet another garbage that it will store in Force.

The FORTRAN fathers thought up the following solution: give the programmers the ability to create “functions”.

If the FORTRAN compiler comes across a function, it will insert a jump to its code wherever it encounters the name of that function. Thus, the following code will work as expected:

Mass = 5.0
Acceleration = 3.0
WRITE (Force)
FUNCTION Force = Mass * Acceleration

In this listing, the computer is still working in a sequential fashion. Yet, it is made to seem as if it wasn’t.

This proved a too-strong an illusion. With the advent of Java and concepts like OOP and operator-overloading, it seemed computers were getting smarter and were learning.

But the truth is, software development tools hit a wall with Visual Basic 5. There has been little improvement since.

It seems software development cannot be made any easier.

This illusion is one reason why scientists have failed to see the problem and the necessity for doing away with the practice of Computer Programming altogether.

I once read in an article about a scientist who manually fed some 2 million facts into a computer. The computer was then given the following question:

A guy entered a fast food restaurant and ordered a burger. What did he eat?

The computer couldn’t answer the question as there was no information provided about what the guy had eaten.

The article writer contended that any 3-year-old could answer such a question.

My niece was 3 years old.

I put the question to her and she answered “a burger”!

Amazing, isn’t it?

Another reason behind scientists’ failure is that “intelligence” is a misnomer.

Computers lack no intelligence and we all know that. This has confused the scientists because they don’t know where to start and what they need to build.

It’s like you want a bigger car to move more passengers at a time but you keep telling the salesman that your (sports) car is not fast enough!

Now I’m going to assume that by this third post, I have presented enough evidence & apology for my belief that traditional approach to AI is totally misguided.

Beginning from the next post, I’ll start presenting my approach to AI: machines that are capable of behaving like humans.

I like the name Terminator for such machines because T-800 was an evil device in 1984 that was mysogynist and sought to kill a woman named Sarah Connor and it failed.

But by 1991, it had a change of heart and became a protector of women & children. This is what Sarah Connor thought about it in 1991:

“Of all the would-be fathers who came and went over the years, this thing, this machine was the only one that measured up.”

Till next post, see ya.

Terminator Genesis 1 – The Need For Terminators