Sultan A. Alshaali is an expert executive in driving growth with 18 years of experience in both the public and private sectors, and nonprofit NGO's leveraging organisational strengths and available resources.
- The problem: There are so many misconceptions about artificial intelligence that glorify A.I. as the ultimate solution to most problems confronting us in the digital age.
- Why It matters: It sets up false and inaccurate expectations and premises that may lead to catastrophic results.
- The solution: Addressing the underlying issues and misconceptions, as well as preventing the myths that glorify artificial intelligence.
Companies position themselves as "bleeding edge" technology users. They identify as trendsetters and leaders, better than their competition and better than you could expect to find anywhere else. It's all part of the game.
The CEOs, Advertisers, and Marketing Divisions pushing Artificial Intelligence (A.I.) are almost certainly not technologists suited to identify the single most complex endeavour in human history.
They promise us that A.I. will:
- Make you immune to hackers;
- Improve customer service;
- Eliminate repetitive tasks, freeing staff to help customers;
- Enhance product design;
- Predict customer needs and desires;
- Evolve your business from local to international;
- Leverage your Big Data;
- Advertise one-to-one, customised for each client.
Except it's not true.
Machines are neither benevolent or malevolent—they exist—and any morality they have will be what we programme into them.
A.I. is not driving our technology advancements; it's happening due to information gathering, sorting, and manipulating data with algorithms—not Artificial Intelligence.
We're on the cusp of a new age, but we have yet to determine when it will mature. Enthusiasts imagine the wonders which await us, like the brief list above, but even they are severely underestimating the potential.Â
Others, like Stephen Hawking (an undeniable mathematical genius), are stepping outside their field of knowledge like Elon Musk and many others. They warn us about "The Singularity"—the Rise of the Machines—Humanity's Doom.
Famed scientist and science fiction writer Isaac Asimov once described "The Frankenstein Complex" that pervaded early Science Fiction. There was a notion that if humanity were to tread on the toes of Gods by creating life or investigating the universe, we would be slapped down and punished for this hubris. He said it was patently silly—and he was right.
The only constraints placed on humanity are the ones we put there ourselves. Machines are neither benevolent or malevolent—they exist—and any morality they have will be what we programme into them.
Failures lead to success
At the 2019 Turing Talk on AI, one expert (Dr Krishna Gummadi) said machines cannot handle emotional intelligence. There is no hint that non-biologics can support emotion of any kind.
If possible, we're almost certainly decades (maybe centuries) away from such a development. Human brains have the thalamus and the amygdala buried deep inside. These tell us what we are feeling and how we feel about it.
But can you programme a computer to "be happy"? How about jealousy, sadness, anger, or disappointment? There is no flood of chemicals, like within you, to make you ball up your fists and start a bar fight, cling desperately to a lover, or cry at the death of your puppy. We cannot do these things for machines, and it wouldn't be desirable in any case.
Instead, we must train our models with pure, untainted ethics. That is what they require to function, and they will work just fine. Poor-grade Science Fiction would have us fear Artificial Intelligence modules taking over or merely finding them irrelevant. We're curious if making them care about us, themselves, or anything else is possible. But if we design them to be ethical, that problem goes away.
Ethics is a serious issue
In an embarrassing experiment designed to eliminate the U.S. Justice System's inherent bias against people with dark-coloured skin, an A.I. programme was designed to help judges create fairer sentences. It reviewed thousands of cases to learn how sentencing worked, understand recidivism rates, and who was most likely to reoffend. Unfortunately, it then continued to recommend inappropriately harsh sentencing for dark-skinned people.
The researchers removed racial references from the data. Still, the algorithm had already formed conclusions about first and last names, low income, specific neighbourhoods, sex, and age, and continued to hand out harsh sentences. In computer terms, we call this GIGO, or Garbage In, Garbage Out.
You must teach an A.I. with correct data. If you don’t, you will perpetuate the problem. If the programme had been brilliant, it would have figured this out and eliminated the problem. However, since "A.I." is still stupid, all it did was figure out how to do what was already being done and continue doing it just as badly.
Reality sets in
Someone may come up with an insight that allows true A.I. very soon, but it is equally possible that it could take ten years or twenty! We're making progress, to be sure, edging our way closer to this incredible goal, but we're not even a substantial fraction of the way towards achieving artificial intelligence. All of the purported A.I. that you currently experience is similar to a clever parrot.Â
Think not? Are you convinced that Alexa, Cortana, Siri, and others are examples of real Artificial Intelligence? Do you think that AMAZON™, Google™, eBay™, and others are using real Artificial Intelligence? Sorry, but no, they are not.
How "humanity" is achieved
The first step in that process is to use a female voice of about 30 years old because it is young enough to sound attractive but sufficiently old enough to engender feelings of maturity, trust, and respect—it's a "Mum" voice—and works well for both men and women.
The second step is the programming-heavy portion, where the algorithms recognise keywords and their relationships. Finally, for the machine, results are placed on a spectrum from "successful" to "poor" and assigned a probability.
The programme finds the best match for your version of the question and parrots the most likely answer, which can happen in milliseconds, creating the illusion of a meaningful conversation.Â
No intelligence involved
We don't need Alan Turing and his Turing Test to see that all A.I.s are profoundly "not human". So, for example, ask Alexa, "Can you pass the Turing Test?" and her response would be , “I don't need to—I'm not pretending to be a human".
Ask the Google Assistant the same question, and "she" says, "I don't mind if you can tell I'm not human. As long as I'm helpful, I'm good!" This frankness sounds like something a natural person would say, but an engineer programmed it, not something created spontaneously by the machine. Even IBM's WATSON supercomputer, the machine that beat the human champions of the game show "Jeopardy!", was not intelligent by any stretch of the imagination.
"WATSON was a costly and labour- intensive associational database, carefully crafted to interpret the subtleties of puns and mixed metaphors and subject matter that reflected typical "answers" the gameshow provided. It's programming then told it to convert these clues into a "question" that revealed the relationship between the clues." The premise was simple, although the execution was costly and complex.
Still, it was a small but significant step to making A.I. possible. Earlier versions, like IBM's Deep Blue, were glorified adding machines, but they made chess much more challenging for humans and resulted in chess programmes on store shelves that could easily beat Deep Blue. Consequent "precursor A.I.s", like ALPHA GO, have made advances in machine learning that will eventually help to reach true A.I.
Where we are today
The truth is that all companies have Big Data—usually in the form of records and information that go unused because they're unorganised and stored in too many diverse locations.
We began with rigid data warehouses that were orderly but accessible via tools like SQL (Sequential Query Language). However, it was easier to see relationships if you were very skilful at phrasing the questions correctly and had the cooperative databases available when you posed the question. In addition, you were obliged to pull in multiple tables to pose a question if you expected a meaningful answer.
Hadoop™ came with the concept of Data Lakes, aggregating data from many disparate sources, such as client records and e-documents, but added image recognition, video, audio, web-scraping, and much more. It was flexible compared to SQL, so the data became more generally useful for human enquiries. However, it is still useless for A.I. to do without a good sorting/interpreting algorithm.
Superficially designed to emulate biological neural networks found in animals and humans, these systems can associate disparate data. Moreover, they are generally very low on "rules", so the programme is free to compare seemingly unrelated things.
These artificial Neural Nets can find subtle relationships because they are unconstrained. It's akin to a scientist daydreaming about making a better underarm deodorant while fiddling with a ballpoint pen and suddenly thinking, "What if I made an oversized ballpoint pen with a big roller?" inventing the roll-on deodorant (which happened in 1952 with Helen Barnett Diserens).
Being free to make unconventional associations makes A.I. a better system. For example, once tied to real A.I., it may connect the knowledge of a veterinarian in Poland with an atomic physicist in France, and a scuba diver in Pakistan, to explain how to make Warp Drive possible. When supplied with the totality of our knowledge, A.I. will find answers before us, but humans can only connect some of the pieces.
Machine learning
Researchers once set up a proto-A.I. programme to work on an old 8-bit video game with no instructions except to operate buttons and accumulate points. They left it overnight, and when they returned in the morning, it had mastered the game and was unbeatable. But, of course, it didn't win; it just followed its rudimentary instructions. The researchers found they could do this with almost any game.
The takeaway
Unless programmed to do so, A.I.s will never want to "Kill all humans!" Barring some fantastic biomechanical innovation, they will probably never "want" or "feel" anything. They can behave ethically, but only if we make an effort to programme them with unbiased, pure ethical standards and paradigms.
Many people characterise current A.I. as a hoax. But, in reality, it is merely a reflection of scientists talking about potential, enthusiasts extrapolating about those possibilities, and then naïve media reporting these predictions as if they were foregone conclusions.
It isn't malice on the part of companies trying to trick you into using their products. But, unfortunately, they need to be more informed as they are often misguided by promoters misappropriating terminology and equating Machine Learning with A.I..Â
Except for a few scientists doing significant research in the area, only some people truly understand what A.I. is. Now you are somewhat more informed than the average person, so spread the word—and though it is possible, it's probably best not to expect the authentic appearance of the first basic signs of A.I. before 2030.
đź‘‹ You can create a post like this one! Share your thoughts with a community of public servants. Learn more
(Image Credit: Unsplash)
Log in or sign up to continue the conversation