From the Google Assistant to self-driving cars, the Fourth Industrial Revolution is progressing rapidly. While science fiction portrays artificial intelligence (AI) as robots with human-like characteristics, AI encompasses anything from IBM’s Watson to autonomous weapons. Although AI and Machine Learning (ML) are household terms, not many people are clear on what they actually mean.
What is Artificial Intelligence?
Less than a decade after breaking the Nazi encryption machine Enigma, mathematician Alan Turing changed history yet again with a simple question – “Can machines think?”
At its core, AI is a branch of computer science that aims to answer Turing’s question. It is an endeavor to replicate and simulate human intelligence in machines. AI is achieved by analyzing how human brains work while solving a problem and then using problem-solving techniques to build complex algorithms to perform similar tasks. It is a decision-making system that continuously learns, adapts, suggests, and takes actions automatically. It requires algorithms that are capable of learning from their experiences, which is where Machine Learning comes into the picture.
ML, a subset of AI, is a field of study that gives computers the ability to learn without being explicitly programmed. ML can be used to solve tough issues like enabling autonomous cars, face detection and recognition, and credit card fraud. ML programs constantly iterate over large data sets, analyzing the patterns in data and facilitating machines to respond to new, different situations. Machines learn from the past to predict the future.
Subsets of AI
The broad definition of AI is vague and can cause misrepresentation of the type of AI that we interact with today. It is useful to define AI in terms of its subsets, namely narrow, general, and super AI.
Narrow AI – also referred to as “applied AI” or “weak AI.” All of today’s AI is narrow. Narrow AI (ANI) is purpose-built and focused on a specific task within the “narrow” context of the product or process. While it appears that today’s machines can do everything, they actually focus on doing one particular thing well. These systems would be useless in pursuits beyond those for which they are specifically designed – for instance, your Google Maps GPS wouldn’t be able to tell you how many calories your donut contains.
General AI – also referred to as “strong AI” or AGI. It is the pursuit of a machine that has the same general intelligence as a human. For example, just as humans can discuss politics, tell a joke, and then play a sport in the span of a few moments, the AGI computer will have the general intelligence to perform these activities as well.
Creating AGI is much harder than creating ANI; by most estimates, we are still at least two decades away from developing such AI capabilities.
Super AI – ASI will surpass human intelligence in all aspects – from creativity to general wisdom to problem-solving. Machines will be capable of exhibiting intelligence that even the brightest humans can’t imagine. Elon Musk believes that ASI will lead to the extinction of the human race.
Yet, like Andrew Ng, chief scientists at Baidu Research, put it, “worrying about general or super AI is like worrying about overpopulation on Mars before we’ve even set foot on it.”
The Future of Jobs
With the advent of AI and ML, numerous tasks and jobs will become automated. Oxford University estimates that 47% of United States jobs will be automated by 2025. However, the math doesn’t add up, since this means that 173 million jobs will be eliminated via the new technology wave.
A more realistic estimate is that 12% of jobs will be lost to automation in the next 10 to 15 years. While 19 million jobs will be lost, it is predicted that there will be around 21 million new jobs created as a direct result of this Industrial Revolution. If unemployment rates stay constant until 2025, AI is expected to change the labor force in three ways:
- Job Automation: 12% of existing jobs will be at risk of being taken over by modern automated systems
- Job Enhancement: 75% of existing jobs will be enhanced or changed by the new technology, leading to higher outputs for all current jobs
- Job Creation: 13% of net new jobs will be created as technology creates new revenue opportunities. What these jobs entail is impossible to predict and imagine
In the grand scheme of things, only specific aspects of jobs will be automated. Jobs themselves will be augmented by the introduction of automation technology.
Pros and Cons of AI
The discussions about the importance of AI in our lives has gained momentum in recent years. Is it a boon or a bane to the future of human existence? Like all things, AI comes with myriad advantages and disadvantages.
- Reduction in Human Error – humans make mistakes from time to time. Computers, however, do not make these mistakes, provided they are correctly programmed. With AI, the decisions are taken from the previously gathered information and applying a particular set of algorithms. So, errors are reduced, and there is higher accuracy.
- Available 24×7 – An average human works 5-6 hours a day, excluding breaks. Humans require breaks and time for themselves. But using AI, we can make machines work 24×7 at maximum efficiency.
- Helping in Repetitive Jobs – In day-to-day work, humans perform many repetitive tasks, like verifying documents for errors. Using AI, we can productively automate these mundane tasks, increasing human creativity.
- Faster Decisions – Using AI alongside other technologies, we can make machines take decisions faster than a human and carry out actions quicker. While making a decision, humans will analyze many factors, both emotional and practical. AI-powered machines work on what they are programmed to do and deliver the results faster.
- Medical Applications – One of the most significant advantages of AI is utilized in the field of medicine. Doctors and physicians assess the patient’s health-related data and intimate the risk factors to the customers via the health care devices with the help of machines. Also, maximizing AI can lower health care costs and significantly increase the number of people receiving care.
- High Costs of Creation – As AI is updating every day, the hardware and software need to get updated with time to meet the latest requirements. Computers need repair and maintenance, which is very costly. Further, creating these machines is expensive since they are incredibly complex.
- Lacking Out of the Box Thinking – Machines can’t be creative. Since all of today’s AI machines are narrow, they can only do what they are being taught or commanded. Although they help in designing and creating, they aren’t yet at the level of a human brain. Humans are highly sensitive and emotional. We can see, hear, think, and feel. Our thoughts are guided by our feelings, which machines lack. This can prevent sympathizing with emotions for human contact, which prevents machines from working effectively in professions such as nursing, for instance.
- No Human Replication – In the foreseeable future, no matter how smart a machine becomes, it can never replicate a human. Machines are rational, but they don’t possess emotions and moral values. Therefore, they don’t know what is ethical and legal. They do what they are told, and hence don’t have judgment making skills. They fail upon encountering new situations.
- Loss of control – As seen partially with smartphones and other technology already, humans can become too dependent on AI, and we begin to lose our mental capabilities. We tend to get addicted to these new inventions, which can cause problems for future generations.
In recent times, the debates about the pros and cons of AI have intensified. But one thing is certain – AI isn’t coming; it’s here. Rather than fighting this or trying to predict the future, we need to go out and work hand-in-hand with the new machines.
Sources: What To Do When Machines Do Everything – Malcolm Frank, Paul Roehrig, and Ben Pring, general research on the Internet