Table of Contents
What is artificial intelligence?
“Artificial intelligence” (AI) is the science of making computers do things that require intelligence when done by humans. AI is a broad field, encompassing many subfields like machine learning, deep learning, and computer vision. AI can be used to solve a wide range of problems (for example, recognising speech or playing board games), and it has been used in several fields (including medical diagnosis, military simulation, stock market trading). In this article, we will explore the different aspects of artificial intelligence and what you need to know.
AI is sometimes treated as a superset of machine learning: while machine learning focuses on teaching computers to learn from data without human intervention and make predictions about the future based on past events, AI goes beyond machine learning and covers broader areas of problem-solving such as logic-based algorithms and search. In this course we will often speak about “machine learning” when we mean “AI”; but in other contexts, you may encounter people who use these terms differently to describe more specific problems.
How might AI affect the economy?
The current AI boom is generating a lot of hype. Many experts say that AI is coming, and it will change the world. The trend appears to be growing more popular by the day, so we thought it was an appropriate time to address some of the concerns about AI in this introductory guide. Sure, it’s not something you can fearfully ignore or dismiss at a casual glance; AI is making its way into our daily lives and affecting our work and leisure activities in ways we may not even realise yet.
As for how it might affect the economy, many experts predict that AI will create new types of jobs and also eliminate many existing jobs. While there’s no arguing that emerging technologies are creating new economic opportunities in areas like data analysis, machine learning, robotics, self-driving cars, bioengineering and so on—the end result could be maybe a net loss of jobs overall.
If a computer can do my job, will I lose my job to a machine?
If a computer can do your job, will you lose it to a machine?
Unemployment’s been a concern for the workforce since the Industrial Revolution. But now, with machines replacing humans in so many jobs, from fast-food workers to lawyers and journalists, we seem to be on the cusp of an automation crisis that threatens to leave millions jobless. What will happen to society if computers are able to do human work better and cheaper than people can?
This question is at the heart of one of today’s most worrisome debates about artificial intelligence. To some observers, AI’s disruptive potential means that we face an inevitable decline in employment in the coming decades. It is hard to argue with this kind of pessimism given recent advances; algorithms have indeed been taking over tasks once thought safe from automation: writing stories based on business figures (Bloomberg), diagnosing eye diseases (Google), and translating languages (DeepL). Even more worrying is that these systems are often trained using data from existing products or services—further contributing to their exponential growth as they get smarter and smarter over time.
How might artificial intelligence change society?
It’s too early to say exactly how AI will change our everyday lives over the next several decades. Even though we have a good understanding of how the technology works and what it can do now, many unexpected developments are likely to crop up.
But there are a few things that you can expect if AI continues on its current trajectory:
- You’ll be able to accomplish more in your job and create additional value for your employer or your customers. This is particularly true for knowledge workers, such as data analysts, doctors, accountants, lawyers and marketers.
- Your kids may spend less time in school than you did at their age because they won’t need to learn as much from their teachers and textbooks before moving on to more difficult subjects. Instead of doing long division by hand when they’re seven years old (like I did), they might use AI software that can solve any arithmetic problem instantly by the time they reach kindergarten. There will still be plenty for teachers to do—just perhaps not quite as much book-learning as before.
- You could get better health care than ever before because AI systems might reduce errors and improve diagnostic accuracy compared with traditional human doctors. At the same time, fewer physicians will be needed since many routine tasks like reading X-rays or diagnosing common diseases could be automated using AI technologies once they become sufficiently sophisticated (which may take longer than many people think).
What does AI have to do with machine learning and data science?
Let’s start by distinguishing data science from machine learning and AI. You may have heard the terms data science, machine learning, and artificial intelligence used interchangeably in the tech world, but there are differences among them. The field of data science is an umbrella term that can include AI and machine learning methods as well as a host of other techniques for extracting knowledge from data, such as statistics and visualisation. However, in most cases when people use the term “data science,” they’re referring to a subset of techniques that uses machine learning to analyse millions or billions of pieces of information about individuals—what a person does on their phone, where they go online, what purchases they make—to predict what will engage them next.
Data scientists work on projects such as improving search results on Google Maps or making YouTube recommendations more relevant (and therefore keeping you watching). Data scientists are commonly found at large companies like Google and Facebook (or in smaller divisions within these companies) or at startups funded by capital raised through venture capitalists; their skills are also prized by industries like retail or online dating.
Will AI replace humans in the workplace, or will it supplement human capacity?
The short answer is—both. Many jobs will be lost to AI, and many jobs will be created because of AI. However, these benefits will not be evenly distributed; some people and places will benefit more than others.
Over the last few decades, there’s been a shift in the workplace towards more creative, cognitive-based work that requires problem-solving and greater social skills. While this trend is likely to continue, so is the increasing automation of routine tasks by machines. Overall, there are likely to be more job opportunities for those who can work with machines rather than those whose work can be automated by them.
AI is already quite prominent in the workplace, but it’s not totally clear whether AI will replace human jobs or supplement their capacity. The answer depends on what you mean by “replace.” For example, there are more than 250,000 ATMs in the U.S., and humans have not yet been replaced for the job of providing access to people’s money. However, most people do get their money from ATMs instead of bank tellers these days. So you could say that ATMs have replaced tellers in a way, but that doesn’t mean that the number of teller positions has gone down—it just means that they’re doing something different now.
This is why we believe that AI will mostly supplement humans’ capacity to work rather than replace them entirely. People who work with computers all day long can read hundreds of e-mails as quickly as a computer can read them, and they can interpret the data just as well if not better than a machine can so far. AI will help us by making our jobs easier, faster, and less error-prone—but it won’t be taking the jobs away from us anytime soon.
Is there an alternative to using personal data for advancing science and commercial development?
Many people are concerned about the use of their personal data. Is there an alternative to using personal data for advancing science and commercial development?
In a word, yes. It is possible to anonymise the names of individuals in a dataset and still use that dataset to effectively train an algorithm to see patterns in healthcare data to offer faster diagnosis to patients. The same goes for many other types of information, such as spending records and driving habits. This type of data is called anonymous because it is not tied to a specific person.
If your name is stripped off your financial information—or perhaps even just your social security number—then the information still contains statistical value but not personally identifiable information (PII) (per GDPR standards). And as we discussed in chapter 3, machine learning techniques are perfectly well suited to finding patterns in anonymous data. If a company wants to analyse how you shop but you don’t want them to know exactly who you are, they can use machine learning on anonymised transaction data. And if health researchers want to learn more about disease progression without having access to any personal information about individual patients, they can do so by performing mathematical analyses on anonymous medical records or by analysing tissue samples that have had identifying information removed from them (commonly referred to as de-identified).
What if my company adopts AI and automates part of my job away?
Let’s say your company decides to adopt AI and automates part of your job away. Should you be worried? Over the long term, probably not. In many interviews with CEOs that we have seen, they were unanimous in their belief that overall, AI will create more jobs than it takes away.
If some of your tasks are automated by a machine, then you should look for a new job—but don’t panic; others will come along. If you work in manufacturing, explore opportunities in other industries or other parts of your company; if you’re an accountant or financial analyst, investigate whether there is something else in your company that needs doing. You may need to up-skill yourself—or perhaps re-skill yourself entirely—in order to take on these new challenges.
Artificial intelligence is here, and it’s changing the world.
Have you ever received a highly personalised recommendation from your favourite retailer, based on a recent purchase? That’s also thanks to Machine Learning, which can be considered a subset of AI (or artificial intelligence).
In the past five years alone, we’ve seen AI become an everyday part of our lives. Some of us may be aware of this; others may not have even realised it until now. But just think: how many times have you used Siri or Alexa in the last day? How many times did you check Google Maps to find the best route to work? What about Snapchat filters and Facebook suggestions—are they starting to feel less like simple tools and more like common sense?
However, only time will tell how AI impacts our lives in the future, and whether it pushes us towards a utopia or more of a dystopia. Our only hope is that we will be prepared for whatever the future may hold.