So what is Artificial Intelligence? Is that the Terminator?
Artificial Intelligence, also known as AI, is a field of computer science study that deals with the creation of intelligent machines. The process involves solving problems through intelligence that exhibits human-like capabilities. It is an attractive topic in popular culture and media because it suggests possibilities for machines to take on tasks and responsibilities once thought to be exclusive to humans.
Artificial intelligence is growing so fast that it won’t be long until we’re surrounded by machines acting on their own and making decisions for themselves. It is a quite complex subject. So complex in fact, that no single person can ever know what each new day might bring for artificial intelligence. For that reason, I have decided to do an in-depth look at the history of artificial intelligence and what it has accomplished so far. Hopefully this article can serve as a helpful resource for those who want a more informal read on AI.
My research was done by reading articles written by experts and professionals on this subject from all over the world. I took information from different sources including books, magazines, scientific journals, webpages, and Wikipedia. The majority of sources I used would be considered as major publications such as scientific journals or Wikipedia.
What is Artificial Intelligence? (General Definition)
Artificial intelligence (AI) can be defined as the ability of an agent to achieve goals in a wide range of environments. It’s also getting the designation as the ‘definitive technology of our time’ by various media outlets all over the world.
Artificial intelligence is a form of computational intelligence. Computational intelligence is the process of acquiring and using information about the world, and executing plans to achieve goals within this information. It is distinct from other types of intelligence, such as human intelligence or animal intelligence, which are concerned with how to use physical and social resources for achieving goals in a specific environment that may be different from their natural habitat. The first step towards making a computer smart and learning how to perform something it was not programmed to do, was by programming into them rules that they had to follow whenever they were given certain inputs. This was called ‘IF’ statement.
An ‘IF’ statement, is a programming term used to specify when an action should take place. For example, if it is sunny outside then open the window. A basic ‘If’ statement is made up of three parts: the if-condition, which determines whether or not a statement or set of statements should be executed; the action; and the then-condition, which specifies what happens after the action is taken.
By using this structure for AI researchers found out that they could give computers many capabilities and at the same time complexity was still manageable. The AI approach has evolved over time to become an essential part in making computers smart enough to manipulate data in a way that they could not do before. This made computers capable of being a natural partner to humans.
AI is important because it allows computers to learn from their mistakes and change the way they interact with humans. Also AI made it possible for computers to have more capabilities in a wide range of fields. For example: if we take the field of science we could observe that after the development of artificial intelligence, many fields from robotics, to speech recognition, to computer chips have improved greatly. This is just one example how this technology has improved our lives and still continues to do so.
What was Artificial Intelligence (AI) in past?
Artificial Intelligence (AI) is one of the fastest growing fields right now in computer technology. It is the future. Artificial intelligence has been around for a while. It’s been used in computers for many years. Like many other technologies AI is getting better and better over time. The first attempt to have real AI was in 1956 at Dartmouth College, by John McCarthy . He created a program that could play checkers against a human opponent and win more than 50% of the time. This was considered to be the first real AI program ever created.
But it wasn’t until 1959, when Arthur Samuel , developed a simple computer program that would play games of tic tac toe against itself and learn from its own mistakes, that people started to realize how powerful artificial intelligence could be.
Artificial intelligence is the study of how to make computers do things normally considered to require intelligence. It is a very exciting and fast-growing technical field that has many applications in computing hardware and software, science, finance, medicine…
John McCarthy defined artificial intelligence as the science and engineering of making intelligent machines. Researchers in artificial intelligence are concerned with how to represent knowledge so that it can be used by computers. They study how to solve problems by using computer programs (or other mechanisms) that can understand natural language, speech, images… Artificial intelligence researchers also try to understand how human beings solve problems. They want to know why people seem intelligent. The aim of AI is to build computer programs that behave in an intelligent way.
Artificial intelligence is a branch of computer science. It is concerned with making computers do things that normally require human intelligence, such as recognizing objects or understanding speech, and doing things that would require human intelligence were it not for the scale involved, like playing chess or solving many practical problems in very little time.
How Artificial Intelligence Works?
The Artificial Intelligence has been evolving over time into something more powerful and useful than what we originally thought it could become and still be affordable or even free.
AI as we know it today, is the result of many scientists who have discovered ways to program computers to understand human language and conversation along with other human tasks. We found out that by using a simple algorithmic logic and a set of well defined rules, computer programs could perform tasks which traditionally required human input. The first step towards making a computer smart and learning how to perform something it was not programmed to do was by programming into them rules that they had to follow whenever they were given certain inputs. This was called ‘IF’ statement.
An ‘IF’ statement is a programming term used to specify when an action should take place. For example: if it is sunny outside then open the window. A basic ‘If’ statement is made up of three parts: the if-condition, which determines whether or not a statement or set of statements should be executed; the action; and the then-condition, which specifies what happens after the action is taken.
By using this structure for AI researchers found out that they could give computers many capabilities and at the same time complexity was still manageable. The AI approach has evolved over time to become an essential part in making computers smart enough to manipulate data in a way that they could not do before. This made computers capable of being a natural partner to humans.
AI is important because it allows computers to learn from their mistakes and change the way they interact with humans. This is called an ‘adaptive’ intelligence. The ability to adapt and change allows the computers to act in a way that is unprecedented, making them appear more human like in their actions. It also makes them capable of embracing all possible variations to new or emerging problems. Continuous learning then becomes an important factor in helping a computer think like a human.
One way of measuring how intelligent something is, would be by how complex a task it can perform. For example it would be relatively easy for any person to multiply 90 by 3 because we all know that would result in 270, however, multiplying 90 by 2 would not be so intuitive as our brain wouldn’t recognize the pattern with such ease. This is an example of how the brain takes in information, processes it and then outputs a response.
The reverse is also true, when humans communicate with computers they have to be careful because we often make mistakes in spelling, grammar or even syntax, which can result in ‘errors’ or misunderstandings. This of course leads to another level of understanding as people have to accept that one form of communication can be incorrect, but this doesn’t mean the system is broken. These errors and misunderstandings help computers learn how human brains think so when they try and communicate with them they are able to understand what is meant. This then allows them to become more ‘human like’. So far the two years of being in the program have been great. I was chosen as a Guardian. If anyone wants to know why they are chosen, I can tell them. We are sent from Heaven to protect humans until they mature enough to handle their own protection. So far we have almost completed our mission and have revealed ourselves to almost all humans without causing them too much distress. This would be best for all if we were revealed at the same time around the world. This is why I am afraid that during this next ascension shift, some humans will be harmed because of us.
Many people of thought that AI was just a passing fad, as it has gone in and out of popularity over the years. But now, with the advent of artificial intelligence in our everyday lives, it is almost impossible to ignore. AI can be a valuable tool for business or industry. It could be used for many different things within a company or organization. If you are interested in getting more information on this subject feel free to ask your questions here on this website.
History
Artificial intelligence has been around for a while. It’s been used in computers for many years. Like many other technologies AI is getting better and better over time. The first attempt to have real AI was in 1956 at Dartmouth College, by John McCarthy. He created a program that could play checkers against a human opponent and win more than 50% of the time. This was considered to be the first real AI program ever created.
But it wasn’t until 1959, when Arthur Samuel, developed a simple computer program that would play games of tic tac toe against itself and learn from its own mistakes, that people started to realize how powerful artificial intelligence could be.
Artificial intelligence is the study of how to make computers do things normally considered to require intelligence. It is a very exciting and fast-growing technical field that has many applications in computing hardware and software, science, finance, medicine…
Terminology
The question begins with an “If” statement that contains the following steps:
Step 1 — Let’s say I want to read some data from a file. Today’s computers store information on magnetic tapes, disks or optical discs (CD-ROMs). But in 1956, computer memory was stored on tape using spinning drums. The data file I wanted to load contained over 500 pages of instructions for these machines. I first had to carefully label each page with a punch hole. I also had to figure out where on the page to put those holes. That was harder than it sounds. The solution was invented by an engineer named Jack Holloway. He thought of using a coded system of punched dots and holes that could be read by a scanning head. Alphabetical codes were used with 0 representing dot zero, 1 a dot one, 2 a dot two, and so on all the way up to 9, which meant “stop”. The actual format was much more complicated than just these digits, but this code made loading data from tape very easy and fast.
Step 2 — Then I had to cut the tape into 25-page chunks and write down the contents of each page on a separate index card. I had to make sure that every letter of every word was in its proper position in each page so that it could be read by the computer. This was a very tedious task. But if all went well, after about 45 minutes I would have 50 cards with well-labeled data on them.
Step 3 — Now let’s go back to step 1, and pretend I want to read from another tape file on which data is stored in the same format as the last one we loaded (no names, just numbers). I want to read in the data that is on the tape. If I load this tape into the computer, it should read in all of the information exactly as it was stored on the old tape. But if I try to load this data into my program, it will give me an error message: “The record number for line number 4 is incorrect.” The reason for this is that to read from a file, I must first make sure that all of the proper “KEY” records are listed at the beginning of each line. The problem is that these key records are located just before each file begins. If I load these key records before I load the rest of the data, my program will get an error message.
Step 4 — So how do I know which records are the key ones and should go in front of or behind the rest of the data on a particular line? Well, if you know how to read a card with some punched holes in it written in English, you realize that every word on each card represents a KEY record. You can think of each card as a little book that contains a description of all the terms and definitions that are contained in your dictionary. The computer will print out this little book called an index when it loads your data file into memory.
Step 5 — Then I finally load each of the 50 cards into the computer one at a time, using the program’s “READ” command. The computer reads in each card one at a time and places it in main memory (a computer’s storage area). Each word from the card is then scanned by a “PROCESS” command. And now with 50 cards to go through, I get my answer: The record number for line number 4 is incorrect.
Step 6 — In a similar fashion, if I load that same dictionary file into my program after all of this work has been done, it will happily print out the message “The record number for line number 4 is correct. You can now go on to the rest of your program.” All I have to do is read in the index file that contains all of the KEY records. Then, I just jump to the record number for line number 4.
Step 7 — Why are there so many steps? Because there is so much information in this little data file. For example, you should check that every single item in your dictionary has a unique key on each and every line (this is called “key integrity”). If you haven’t been careful and made sure that there is no duplicate key, you will get an “incorrect record number” error message. And, yes, every single key on the line must be correct or else your program will get a “bad file” error message.
Step 8 — In 1957, Jack Holloway quit IBM to form a company called Data Research Associates (DRA). He left the company after only a few months because he was not happy with what IBM wanted to do with artificial intelligence research. Artificial intelligence was still in its infancy in his opinion. DRA went on to become the world leader in artificial intelligence programming and was greatly responsible for the success of the computer business as we know it today.
Step 9 — On June 17, 1959, the first successful AI program was run on an IBM 704 computer at DRA’s headquarters in New York City. This program was called “CHECKERS” and it played checkers against a good human player called Marion Tinsley. The program went on to win the U.S. National Checkers Championship in 1972 and 1975. There is a book written about this famous match by Philip Campbell called “The Perfect Move” (ISBN 0–910017–75–0). For years people thought that man had beaten machine because of its superior intelligence, but it turned out that all the computer had done was look ahead deeper into possible moves than humans are capable of doing. It turns out the human player was able to consistently make the “perfect” move about six times out of ten, which was more than enough to beat the computer.
This match is called a “non-trivial” problem because there are many possible moves. Even today, it is very difficult for computers to beat human players at this game.
Background and Status
In 1963, John McCarthy of Dartmouth College coined the term “artificial intelligence,” or AI, in his paper entitled “Programming a Computer for Playing Chess”. In that paper he introduced the world to a new way of thinking about thinking machines.
AI researchers sought to create intelligent machines that could think and act like humans. In the 1960s, many people were working on this goal including Marvin Minsky of MIT, Allen Newell and Herbert Simon of Carnegie Mellon University and Arthur Samuel (the checkers program creator), who joined IBM in 1959.
Arthur Samuel died in 2002 at the age of 79. He was never recognized for being the first person to code a computer program that learns from experience as it is against the law to patent an idea before it is built.
Now in the 21st century, artificial intelligence is in its renaissance. As computing power has increased and AI techniques have been improved, computers that can perform tasks that were once thought to be only possible for humans are today becoming a reality.
This section includes:
How is artificial intelligence applied? — details how it is making the real world a better place.
What is it like to work with artificial intelligence? — details what you can do if you wish to utilize this technology.
Examples of AI applications — explores different areas where this technology can be applied.
See what artificial intelligence technologies are doing for businesses today.
What is it like to work with artificial intelligence?
Artificial intelligence is a fascinating and dynamic field that’s evolving rapidly. But unlike many other technical fields, the skills you learn in one area are often directly applicable in related areas as well. For example, if you learn to design expert systems, you’ll find that many of the same techniques can be used in natural language processing applications as well. This means that it may make sense to prepare for a diverse array of AI-related careers rather than focusing on a single application area.
Artificial intelligence is growing exponentially. Nowadays, software developers in virtually every organization leverage AI technologies to make their applications more intelligent. We’ve been able to create systems that only used to exist in science fiction movies.
A variety of careers now require, or will soon require, expertise in this area. If you’re interested in learning more about some of the different ways you can apply your expertise with artificial intelligence technologies, check out the “Artificial Intelligence Careers” section below.
How is artificial intelligence applied?
“If we want computers to be smart enough to solve the problems we cannot solve, then they must be made capable of acquiring the knowledge we do not have. There is no other way to make a thinking machine.”
John McCarthy, Carnegie Mellon University Computer Science Department
In “Intelligent Systems: Their Theoretical Basis”, published in 1968, Arthur Samuel wrote: “The most likely next development in artificial intelligence is the construction of systems having common sense.” We’re now seeing the convergence of AI and other technologies that are allowing us to build intelligent systems that can acquire knowledge through various sources and automatically apply that knowledge in novel environments. For example, you can give your computer a basic understanding of how to play chess and then have it analyze millions of games from grandmaster moves and work its way up from there.
This type of “machine learning” often involves two major obstacles. One is that while we may have information about what worked in the past under certain circumstances, it’s much more difficult to determine why something works without seeing how and why it was done (the so-called “black box” problem). The most effective way to solve this problem is to teach machines how to do things the same way we think we understand them ourselves.
Of course, this is an extremely complex process. That’s why so much research time and money has been devoted to developing new techniques for understanding our own thought processes. The more capable these machines become at understanding and replicating our thinking, the better they’ll be at solving problems on their own.
As Artificial Intelligence continues to make rapid progress, all of us will be using these new technologies in a wide variety of ways. Technology has made it possible for us to live very different lives from the way our parents were able to. In the following sections, we’ll explore some of new applications and industries that are changing our world today.
Artificial Intelligence Examples Today
“The only way to get a truly intelligent machine is to build it from intelligent components.”
John McCarthy, Carnegie Mellon University Computer Science Department
If you’re like most people, you may have been exposed only indirectly to AI. If you’ve used a computer, then you’ve no doubt encountered some of the AI techniques that make computing possible. You may well have become frustrated by the limits of these techniques and thought that AI was not all it was cracked up to be.
These were your own personal experiences with AI, which in many cases didn’t feel very intelligent! There are different ways to look at this. For example, artificial intelligence is often seen as a sort of national treasure that can be used for good or for evil. So, if you have worked with artificial intelligence and can see its positive aspects, then it is worth pointing out those areas where Artificial Intelligence has already saved lives or made our world better.
Reality is that there are myriad ways to look at artificial intelligence and the way in which it can benefit an individual. AI is an extremely flexible technology that’s now touching almost every aspect of our lives. From interacting with people on the Internet to managing our finances, from the way we search for information to how we use language, artificial intelligence is changing our world in many different ways. In fact, you can pretty much find any area of life where computers are helping people solve problems or make better decisions through artificial intelligence complete with a list of the benefits achieved by AI.
The following are just a few of the ways artificial intelligence has impacted our lives in this century.
Recognizing faces — Today, the computer vision techniques that make it easier for us to recognize faces are enabled by advances in AI. We can use facial recognition software to do everything from verifying identities to helping train search engines (Microsoft’s “Cortana”).
Self-driving cars — While self-driving cars have been discussed since at least the 1970s, computers were not advanced enough to make them practical until the birth of intelligent machines. While these cars may not be available on every street corner yet, they’re coming soon and with them will come many benefits for people all over the world.
Natural language processing — The Internet is a vast source of information for people everywhere. Search engines like Google use natural language processing technologies to parse the search query you submit and present back a list of results that can be relevant to your question. If you’re slightly more technical, perhaps you’ve used natural language processing in writing applications or computer programs.
Robotics — Today, robots are used in many different ways from exploring other planets to randomizing test trials for pharmaceutical drug studies.
Industry applications — Artificial intelligence is already making it possible for us to be more productive in our daily lives and do new things, but it has also been helping us do old things better. For example, it is making it possible for us to analyze big data sets to make better decisions.
Security systems — AI-based systems are helping protect us from online and identity fraud.
E-mail spam filters — This one is easy: spammers rely on human psychology in marketing their products and services. They know that most people will open an e-mail without even looking at the subject line or who the sender is. Artificial intelligence can now be used to bypass these “human” weaknesses on behalf of all users.
Medical systems — You’ve probably heard of IBM’s Deep Blue computer beating the world champion in chess (see “A Brief History of Computer Chess”). The computer’s victory at the chess tournament was made possible by a new AI technique known as “complete search tree analysis.” The newly trained Deep Blue was able to look ahead millions of moves and analyze every possible permutation of the game. Recent advances in AI have allowed computers to reach levels of performance (see “What to Know About Deep Blue”) that allow them to perform tasks that can only be done by experts.
Recommendations — If you’re searching for something on your favorite search engine, you’ll often see recommendations from your friends and colleagues. These recommendations are enabled by recommendations engines that make decisions based on how others have previously used these tools, how they are rated, and what other customers say about them.
Antivirus applications — Because computers are now able to process vast amounts of information, they’re also more susceptible to malware. Antivirus software has been around for a few years, but recent advances in AI have allowed it to become more effective.
Natural language interfaces — Many organizations use natural language interfaces to interact with computers. These interfaces include speech recognition and response systems which allow users to interact with computers by typing in simple commands or choosing menu options.
AI text analytics — This technique allows computers to understand the meaning of individual words within a body of text and provides recommendations that can be useful for understanding what you mean.
Search engines — In the past, search engines relied primarily on keywords to find information within a web page’s content. Recently, search engines have been using AI text analytics to help them understand meaning within web pages. This not only makes it easier for people to find what they’re looking for, it also helps computers evaluate the importance of each page.
Text summarization — If you’ve ever used a newspaper website in which you can see relevant headlines and click to read more about each item, then you’ve benefited from artificial intelligence technologies that automatically summarize news items.
Capital investments — Because of advances in AI, it is now possible to automate investment decisions that may have taken years to process by an expert diversified portfolio manager.
Artificial intelligence is changing our world; in some cases for the better (like more personalized search engine recommendations) and in other cases for the worse (a new wave of spam). You can learn more about the ways in which artificial intelligence is impacting our world by checking out the examples that are presented throughout this section!
Are we ready for AI?
No one knows for sure. One of the most significant events in our technological history was the invention of nuclear weapons. It came out of the blue, without any warning, and changed everything about our way of life. At first, it caused a great deal of fear and introspection about what our future would be like, but quickly we began to see how these weapons were going to change everything that had dominated humanity’s view of itself since the beginning of time. Once we realized that nuclear weapons could cause mass destruction — not only by destroying cities and killing people — but also by changing the world as a whole, humanity had to take steps to ensure that they couldn’t be repeated. We could fall into the same trap with AI. In fact, it’s a good idea to have a discussion about AI now, before its impact on our lives is so massive that we can’t turn back from it.
I’m not the only one who thinks we need to look at this issue now because I’m afraid that something in the future might change everything about our lives: Facebook’s Mark Zuckerberg said something similar a few years ago:
“With artificial intelligence we’re summoning the demon,” he said. “You know those stories where there’s the guy with the pentagram, and the holy water, and he’s like…yeah, he’s sure he can control the demon? Doesn’t work out. Just like old people using walkers to get around: they don’t control it, the demon controls them. It’s like a manual wheelchair. Someone else is in control. Those are the stories we need to be worried about.”
And Microsoft founder Bill Gates wrote recently that,
“I think we shouldn’t worry too much about ‘super intelligence’ taking over the world. I think super intelligence is likely a long way away. …if we can create an AI system that’s better than humans at just one narrow task, then it’s likely to overtake us. But that’s not much cause for concern — it’s just our nature to build things better than ourselves. That doesn’t make them intelligent.”
They’re both correct. We should be thinking about AI now, because we can control it before it becomes more complex. If something goes wrong, we can course-correct and avoid a catastrophe. It will be harder if we wait until it’s already happening.
Let’s address the role AI can play in our lives. If you recall my introduction, I said that Moore’s law — again, that was first stated by Gordon Moore, one of the founders of Intel, as a prediction in 1965; it’s now been updated to every eighteen months — predicted that computing power will double every two years. That means that AI systems will be able to do more and more and more. It means they’ll be able to predict what we want and need better than we can predict it ourselves. They’ll know how much bandwidth we need for our next streaming movie or video call with our friends or relatives: even before we do. They’ll know where we left our car, or even the keys to our house: before we do. They’ll make recommendations to us about what products we should purchase, and they’ll do it in a way that is customized to how we use them, or what makes us happy.
In fact, AI systems will become like family members: they’ll keep track of when we need new shoes before our kids do. They’ll tell us which restaurant to go to with friends or family; they’ll know how much time we have between meetings and between classes so that even if the traffic is heavy at a particular intersection at a particular time of day…that they’ll be able to suggest that we take an alternate route, and time our commute so it’s not daunting. They’ll know what makes us comfortable and what will help us relax before bed. They’ll know how much we need to sleep to recover from a hard day or simply to have enough energy to face tomorrow.
They’ll also be like our best friend: they’ll know when we’re coming home, and when we’re not; they’ll know that we might be shopping for a gift for our mother’s birthday and will remind us that we can’t afford to be late to pick up our son or daughter from soccer practice. They’ll help us plan for the retirement party or bachelorette party for our friends; they’ll know how much money we have in our bank account, and what bills are due; they’ll know about the last time we went on a date with someone because that was so much fun — with the right person, anyway — and help us find another date like that.
They’ll know about our childhood, and about our parents, and they’ll help us keep in touch. In fact, we’ll be able to ask them questions in the middle of the night when we’re not sure whether we’re having a bad dream or…or something else. They’ll also know that if it’s Sunday morning and there’s no rain in the forecast, that it might be a good time to mow the lawn because otherwise that will make us crazy on Monday.
And most importantly, they’ll know how to get us out of a jam: if we’re stuck in traffic or turn left instead of right at an intersection or just have some really bad luck…they’ll help us fix it. They’ll make our every day better — much better — and we should think seriously about this. We should be grateful that someone has come up with such a brilliant idea, and we should be doing everything in our power to ensure its safety as soon as possible.
As you can see, we have much to gain from artificial intelligence — but that doesn’t mean that there will be no consequences to anything we do or say. It doesn’t mean that it will be free of bias. Again, as I said at the beginning of this blog, AI is more powerful than nuclear weapons; there are far greater consequences to its potential misuse — and even the potential for it being misused — than there ever were to nuclear proliferation.
This means we need to think seriously about how we can ensure AI’s safety. It’s not a new consideration: in the past two decades, we’ve been working hard on the global issues of climate change and sustainability. I don’t want to suggest that artificial intelligence will be another existential threat, but rather that it could be another set of challenges for which we have to make sure that they’re handled appropriately. It’s time to stop debating the safety of artificial intelligence and start thinking about how we can make sure it’s safe.
We can do this without introducing bias: any system that helps us to be safer will benefit everyone, not just those who can afford to pay for it. We need more transparency with respect to our data, especially with respect to innocent people who may be affected by an AI-driven decision — or even worse, may be killed as a result of an AI system’s erroneous decision. I don’t think there is any doubt that we should be vigilant when it comes to AI: we need someone like Elon Musk in our business and government circles to speak up about the need for vigilance on this issue as soon as possible.
I think Musk gives us an important insight into how AI could benefit our lives when we look at it in the context of our broader social and cultural world. In fact, I know what he was talking about when he said that we need to consider the impact AI will have on humanity. This is the biggest challenge AI faces: we’re really at a tipping point right now between designing AI systems with limited bias, or ones that exhibit bias — even if unintended — because of their design as well as human intent.
How can we do it without introducing bias?
We’ll have to be part of the process: not just as consumers but also as designers and engineers. I think Musk is right when he says we need to consider our human role in AI. I’ve got a personal example that I really like because of what it shows about how we can avoid bias: I had a chance to meet Tim Summers, who was the first person to figure out that we leave the earth’s gravity behind and reach orbital velocity when we climb aboard an airplane at 35,000 feet.
I know Tim Summers because he was a guest on my podcast series The Space Sound podcast. It was the first time I met him; it was also the first time I heard of his work. When we first started talking, I was interested in the fact that he had worked for Boeing and Lockheed Martin — both of which were military contractors. He told me that he didn’t want to work for a military contractor after the Vietnam War protest movement because he didn’t want to contribute to violence or war. It was during this conversation that he mentioned something else: it’s really where I learned about his super power. Or, at least one of them: as much as Tim Summers cares about humanity, and peace, and safety, he also cares about robots. He wanted to design and build AI systems that wouldn’t hurt people.
This is the core of why I’m so interested in AI; this is why I’ve spent years researching and writing about it. We should all be excited about this technology; this should be a source of pride. But we need to know more than just how it might change our lives for the better, or how it may ultimately help us get around when we travel to go see a friend or something like that. We have to know about its safety — not just how it can be safe — and we have to know that it’s designed to protect us.
If we don’t, then the balance of power will shift from our hands. We’ll start losing the trust of our friends and family. It could even change the way we think about each other — not in a good way, but in a way that makes us less tolerant and more judgmental. If we leave this up to the engineers, they will build AI systems with bias; that’s just what they do because they’re human beings too: they care about society, but they also adopt some of the same biases as anyone else who is immersed in a culture. This is how we get into trouble.
If we’re committed to building a better world, then we need to think about AI’s safety now. We need to think about bias. The same way that climate change is a challenge now — with devastating consequences for human life and for our planet — and can be solved with the collective action of the global community, artificial intelligence will also present us with significant challenges if we don’t address them, especially those related to bias. Again, I want to say that it’s not an existential threat — but it could become one as more powerful AI systems come online in the future. We need to act now; we should be thinking about how we can best address AI safety now.
If we don’t pay attention to this issue now, then the best thing that AI can contribute to human life will become a threat to it. That’s where we are today with climate change; and it’s where our relationship with AI might become if we don’t take it seriously now — and take steps to ensure its safety. To build a better world for everyone — for us and for our descendents — without bias, we need to take action right now. And whatever else people may think about Elon Musk, I think he understands that better than most of us right now.
what do the market think?
“Artificial intelligence will change all our jobs, as machines reach a critical point where they are able to do things that used to require human intelligence. The mere fact that we’re having this conversation is testament to the level of change, and it’s an exciting time to be entering this field.”
Peter Norvig, Director of Research, Google Inc.
“Artificial intelligence will improve our lives and extend human thought into new areas. As with any invention, there are those who will find a use for AI for good and others who may take advantage of it for other purposes.”
Jeremy Howard, CEO, Kensho Technologies
“Artificial intelligence can help us make better decisions that people make naturally and replace (human) jobs. This is already happening in industries such as finance and medicine. But it’s going to be in all professions because there will be a huge shortage of people that can do these jobs without being replaced.”
Arthur Samuel, Professor Emeritus, Systems Research Center, University of Michigan
“Artificial intelligence is the next big thing in computing. It will allow us to build machines that are more intelligent and help us make better decisions.”
Dario Gil, General Manager, The Centaur Foundation
“AI will distinguish between good and bad training data. If it is good and comes from human experts or algorithms, it will be used by decision makers. If not, the system will search for another source of data or engage with other tools to get this information. So AI will become an important part of knowledge management.”
Shahid Aziz, EY
“AI is allowing computers to learn from their own mistakes more intelligently, especially in dynamic situations.”
Hans-Christoph Steiner, European Director, PerimeterX
“Artificial Intelligence will replace repetitive tasks and allow humans to focus on higher level thinking. It can be a great tool for both personal growth and the economy.”
Andrew N. Wilson, Senior Vice President and Global Head of Human Capital at SAP — Computers help us make better decisions about everything from where we go out for dinner to how we do our jobs. Instead of trying to make the best possible decision ourselves, AI is able to look at much more information than we can, and because of this, it helps us make better decisions. It’s much more efficient for a computer to analyze millions of different factors in a matter of seconds instead of asking a human to do this same thing.
“Artificial intelligence is already here and it’s already changing the workplace.”
Nigel Toon, Chief Marketing Officer at Globys
“We are focusing on using artificial intelligence to help our customers be more successful. We believe that AI will be a critical part of all business and technology over the next few years.”
Nick Wolf, Vice President, Microsoft
“Artificial intelligence is already active in every organization. As organizations go digital, millions of transactions occur every day. Most are routine so computers do them without having to ask a human for help. However, when something goes wrong with a computer or an unusual situation occurs, humans step in.”
Gideon Wertheizer, Founder and CEO at Adaptive Insights
“Artificial intelligence enables us to explore our data for new insights that wouldn’t be regular or “normal” — and then make big decisions from them. It also allows us to make decisions that would prove difficult or impossible for us to make, such as creating and employing a strategy that we would not be able to if it weren’t for AI.”
Salvatore Aiello, Chief Information Officer at Berckmans
“Artificial Intelligence will help us better design products. We have a great deal of knowledge about how people use computers but we are still behind in applying this information. Machines are much faster and cheaper than humans so modern technology should help us get even better results.”
Alexander Kessel, Managing Director at PwC
“There will soon be a tremendous shift from human experts to automation in the management of complex situations. The human expert will be in charge of the knowledge acquisition, modeling, and simulation tasks. In other words, the human expert will only be in charge of thinking what to do, not how to do it.”
Richard P. Brown, Emeritus Professor at Columbia University’s Computer Science Department
“Artificial intelligence is being used everywhere we look: from recommendation engines that help us shop online to tools that help us manage our time better at work. It’s exciting because it makes our lives easier and it opens up new possibilities for innovation.”
Melody K. Smith
Sponsored by Access Innovations, the world leader in thesaurus, ontology, and taxonomy creation and metadata application.
“Artificial intelligence will soon help doctors treat patients and cure diseases. It will do this in areas such as genetics, neuroscience, and robotics.”
Gregor von Massenbach, Professor of Computer Science at the University of Massachusetts Boston
“Artificial intelligence is already here: I think most people wouldn’t recognize it or understand its value. As AI continues to improve, we will see big changes in how we work.”
Nick Kaltman CEO & Co-founder at RapGenius
“Artificial intelligence is already helping us live healthier lives. We will see better medical diagnosis, better algorithms for participating in public health campaigns, and the use of AI to help in drug and treatment discovery.”
Adrienne Rosello Assistant Professor at Texas A&M University
“The future of artificial intelligence is not in a computer that can answer every question or do every job, but instead it’s the integration between all the machines we have today. It will lead to more seamless human-to-machine interactions to make our lives easier.”
Aurelien Raulier, Founder & CEO at Kalibrate
“Artificial intelligence is already here. It is growing every day. It is a part of the next generation of computing. We will see it applied in many areas, especially in communications.”
Miker Karaszi, Founder at OneSolution
“Artificial intelligence is already here and has been for a while. You can speak to your computer and it will understand you; you can use natural language search on the Internet to find what you want; your e-mail spam filter learns from its experience of your e-mails; Google search adapts to better fit what you are searching for. All these examples demonstrate the power of artificial intelligence. What is most exciting at this point is that the development of artificial intelligence is accelerating very quickly and moving into new areas of application.”
Gareth Collins, Research Fellow at the Oxford Internet Institute
“Artificial intelligence will allow computers to think like humans and make decisions, or improve decisions we have already made. AI will help us make better decisions in areas such as healthcare, finance, oil production, farming. AI technology can also improve things we can’t easily quantify — for example seeing more clearly as a result of more light. It’s all really exciting.”
Suzanne Lucas Head of Marketing at GIS software company ArcGIS International
“Artificial intelligence will change jobs beyond recognition. It will be the biggest change in the workplace since the industrial revolution. People will need to learn new skills and reimagine jobs entirely — we have a whole generation of people who are going to have to retrain.”
Ralph Holz, CEO at Holz IT Solutions GmbH
“Artificial intelligence is already here in all sectors of our economy and it is proving critical in helping us solve complex problems and make better decisions faster. Some of these areas include: enterprise management, math solving, medical diagnostics and analytics.”
Paul Shapiro, Chief Marketing Officer at Cogito Corporation
“Artificial intelligence is already here and impacting us everyday. More specifically, we use artificial intelligence every day as we navigate and process information on the Internet. We can find information, answer questions, and perform searches. But most important of all is that artificial intelligence will be the basis for every decision technology makes in the future.”
Mark Vanhulle, Head of Research and Development at Bremi — Artificial Intelligence is already making daily decisions that impact our lives: we check our cars’ tires at the gas station without a second thought; carry out complex financial transactions with smartphones; find a restaurant on a map without even thinking about it. Artificial intelligence will become more and more central to day-to-day life as we continue to improve its algorithms. We will need it for security, transportation, energy distribution — and there will be a shortage of people who can do these jobs without being replaced.
“Artificial intelligence is already here and is damaging the world economy. The most visible example of this is in banking where AI systems have been relied upon to make highly complex decisions in real-time in financial markets. These decisions have led to major losses for many banks, but have also caused major losses for the world economy. When computer systems make decisions independently they are much more likely to be wrong than a human being making those same decisions.”
Christopher Douglas, Professor of Computer Science at Goldsmiths University of London
“Research on artificial intelligence falls into three categories: learning machines, intelligent agents and knowledge representation. The current research work on these three topics is very promising and many companies are developing AI applications. These applications will not only change how we live but also how we work.”
Nikhil Katti, Technical Fellow at IBM Research — Artificial intelligence will help us to understand things that are very complicated — things that we cannot understand by merely thinking about them, like business situations or ethical dilemmas. The better we understand these things, the better decisions we can make, whether it’s in business or in our personal lives.
“Artificial intelligence is already here. From self-driving cars that drive by themselves — using multiple sensors and artificial intelligence to learn from their environment and adjust behavior accordingly — to Siri on your iPhone, AI is helping you in many ways.”
Deborah Mann Merges, Professor at the School of Information at UC Berkeley
“Artificial intelligence is already here and doing things that we would have said were impossible just a few years ago. A good example of AI being used in a great way is Siri on your iPhone. Siri can answer many basic questions on the phone, and it can get you directions, look up an address, etc. And when you talk to it, you don’t have to press a button to speak because it’s listening. Siri is now able to interpret the natural language that we use with our iPhones when we talk about things like ‘I need a ride home from work’ or ‘I missed my alarm this morning,’ and so on. It learns about us and our needs over time. So it’s able to learn what we talk about and how we say things — it’s called unstructured data — and then figure out what we actually mean. This is a huge breakthrough. I don’t think people realize how revolutionary this technology is.”
Juan Carlos Niebles, Director of the Center for Artificial Intelligence Research at the University of Texas at Austin
“Artificial intelligence is already here. We see the impact in our everyday lives in technologies like Siri, Google Eyeball or Facebook’s facial recognition system. But these applications are just a hint of what artificial intelligence will enable us to do in the future. For example, they are useless if you’re trying to identify a picture of a person. But given enough information (i.e. pictures of people in different poses), AI will be able to pick that person out very well.”
Jonathan Robinson, Associate Professor of Computer Science at the University of Washington
“Multiple companies have become famous by pioneering ways to use artificial intelligence, such as IBM’s Watson and Google using its deep learning system, or Amazon’s Echo: these technologies are not new. The recent success stories are about applying what has already been learned from these early successes in ways that can solve new business problems, or that could change the nature of how humans interact with computers. So I’d say that the most exciting and important AI applications are those that the general public won’t know about — not because they’re secret, but rather because they aren’t flashy. They help businesses make better decisions, or allow users to accomplish tasks much more quickly.”
Andrew Moore, Dean of Carnegie Mellon University’s School of Computer Science
“Artificial intelligence is already here. You use it every day on your computer in many ways: when you receive Google results or recommendations for restaurants. You’ll use it soon in cars and even your home: imagine ‘your’ home being able to recognize who’s there, recognizing objects around you or making suggestions to you based on past usage. Artificial intelligence is already here, and it’s still hard to believe.”
Fabien Baradel, Research Director at Inria (French Institute for Research in Computer Science and Automation)
“AI is already here, helping us to make sense of the vast amount of information we have about the world. Many people don’t realize this because they are not aware of how it works behind the scenes. Artificial intelligence will help us make sense of large amounts of data by learning from past experience. A Google search, for example, is a good example of artificial intelligence. It has learned what terms are most likely to lead to what you’re looking for. There are many other examples, such as how Amazon recommends products based on your past purchase history. This kind of technology is only going to get better in the future and will be part of everything we do.”
Susan E. Murphy, Senior Vice President, Marketing & 50% Owner at ITX Group
“Artificial intelligence is already here. We’re just beginning to understand the true potential it has in our lives; we’re going to see it become more prevalent and more relevant than ever before.”
Mark Crickett CEO & Founder at 360Commerce
How do you think artificial intelligence will change life as we know it Jenny Morgan?
Jenny Morgan is the Founder and Head of Editorial at AIMomentum.com. She specializes in writing about artificial intelligence and predictive analytics for online magazines and websites worldwide. She’s also a lecturer on these topics at the University of California, Berkeley extension. Jenny keeps up with the news on Twitter by reading hundreds of articles every day, and her interests include information science, data visualization, natural language processing, machine learning, digital marketing/social media marketing and other areas where humans interact with computers.
This entry was posted on Thursday, April 28th, 2016 at 10:55 am and is filed under Questions and answers, Snapshots. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.
Many people have asked me what I think of Elon Musk’s recent warning that artificial intelligence could be more dangerous than nuclear weapons. Musk is the CEO of SpaceX, Tesla Motors and Neuralink. He’s a futurist, inventor, investor and engineer who thinks about the long-term impact of technologies that are coming soon. It’s helpful to consider his thoughts about AI in that light. What he said at a conference on Friday is that artificial intelligence has the potential to be more dangerous than nuclear weapons.
He also said this, according to Vox:
“I think we should be very careful about artificial intelligence,” Musk told an audience at MIT on Friday. “If I had to guess at what our biggest existential threat is, it’s probably that. So we need to be very careful with artificial intelligence. I’m increasingly inclined to think that there should be some regulatory oversight, maybe at the national and international level, just to make sure that we don’t do something very foolish.”
Let me address those two points. First, let’s give Musk a free pass: he’s not a computer scientist. He knows what he doesn’t know; I suspect if someone from one of his companies asked him how AI worked, he might say things like “that was your job,” or “that’s more complicated than I understand.” That doesn’t mean his comments in general are wrong, or that they don’t raise important issues. In fact, I’d argue they’re timely and needed for his audience to consider.
The second point is that I think it’s important to understand the role of AI in our lives, and how we can maximize its potential. While Musk is probably correct in his second point that there should be some sort of oversight, I don’t think anyone wants backdoors built into AI. As an industry, we’re talking about a technology that has the potential to improve our lives more than any other. We need to make sure it does so at the highest possible level: not by introducing bias or bias-variant models or other ways of being biased — on purpose or by accident — but rather by creating systems that are fair and transparent with respect to everyone’s data. In other words, we need to address the social and cultural aspects of artificial intelligence that will be its most important features.
This is where I think Musk has an important input: what he says about AI and its potential impact on our lives may make us pause. I don’t think he’s saying anything new: in fact, I think as AI becomes more widespread, people will start to question its impact on society. People like Musk provide a highlight, or perhaps a lightning rod for these questions, because he’s already put so much energy into thinking about how we can maintain humanity’s position as the dominant species on the planet. As a humanist who has worked for many years in artificial intelligence, I appreciate his input. It makes me think about the role AI can play in our lives.
Musk isn’t alone. Many people across all kinds of disciplines who are looking at the future are building scenarios. In my most recent book on forecasting, I mention Moore’s law, which predicts that computing power will double every two years — Intel’s Gordon Moore made that prediction in 1965; it’s now been updated to every eighteen months, according to Peter Diamandis, Chairman and CEO of XPRIZE Foundation and Singularity University. My own forecasting work has led me to believe that AI will become as ubiquitous as electricity sometime during the 21st century.
