We’ve seen it in novels, films, and TV; a scientist creates such sophisticated artificial intelligence (AI) that it eventually becomes smarter than humans and tries to take over the world. But can that really happen?
Artificial Intelligence has progressed by leaps and bounds over the years, but there is still something missing – emotion, logic, and reason.
Can Artificial Intelligence Ever Be Smarter Than Humans?
Humans can feel emotions, both simple and complex. A robot cannot, no matter how intelligently it is programmed. Scientists are doing an amazing job integrating computer science, linguistics, psychology, and neuroscience into AI, but it’s still just elaborate mimicking of human behavior. Machines can collect data to help analyze their past actions and alter future actions, but it’s not quite the same as human evolution.
Henning Beck, a German neuroscientist, makes the following comparison between artificial intelligence and human intelligence:
Humans can learn new words after hearing them in context. We can begin implementing these words in our lives because we understand the concept the word is referring to. A computer cannot make that inference. It needs to be programmed to use the new word appropriately.
While it’s true that artificial intelligence is constantly improving, it will never be able to replicate the human brain’s ability to think for itself and adapt our habits.
How AI and Humans Compare
Like robots, our responses to stimuli can be observed to find the root of a problem. If AI is experiencing a performance error, it might be traced back to its programming, or its assembly. Similarly, if humans are “acting out,” there is usually a deeper issue to examine and repair. Fortunately, with humans, tools such as counseling can help us better understand what is causing the behavior and, in a process similar to server patching in robots, humans can undergo “updates” as well.
Differences Between Human Intelligence and Artificial Intelligence
- AI has to be given data to process. The more data a computer has, the better. In contrast, humans can seek out data themselves, and come to conclusions even when there’s uncertainty or myriad variables such as emotion.
- AI processing time may take longer than human processing time, due to the vast amount of data AI can analyze. Humans can sometimes adapt more readily to changes needed to achieve the desired outcome, shortening the processing time.
- Computers can’t deviate from the rules programmed by humans. They are given instructions to follow, and they stick to them. Humans obviously set and break rules as it suits them. Sometimes breaking “rules” simply means we’re exploring new ways to do something, which could result in better technology or better operating systems for vehicles. A computer would never deviate from the rules, even if it meant making advances.
- Computers cannot grasp abstract concepts; they need data. Human brains can come to conclusions based on concepts, or abstract theories.
- Computers use about 200 watts of energy per hour while operating. Human bodies use about 100 watts of energy at rest, and up to 2,000 watts performing certain tasks such as exercising. For a few minutes at a time, a human can sustainably use 200 – 400 watts.
Types of Artificial Intelligence
There are probably more forms of artificial intelligence around you than you realize.
- Internet Bots – These simulate human behavior based on their coding. A bot can be used to respond to keywords used in communication, generating a canned response to help answer questions. For example, a travel company may use a bot on social media to respond to the word “QUOTE” and automatically respond to let the customer know an agent will be with them shortly with specifics. Internet Bots may also be known as spiders or crawlers, which search the web repetitively to index search engines and collect other information. Harmful bots are referred to as Malware bots, and they are programmed to infect other computer systems and take over. Experts estimate that by 2024, the majority of routine work will be automated in some way. More AI will be employed to complete rote tasks that don’t require human interaction. It will not eliminate the need for skilled labor, but it can increase efficiency.
- Humanoids – A humanoid is a robot built to look like a human form. These types of AI may be sued for research purposes, or performing tasks less safe for humans. You may be familiar with Sophia, a humanoid ambassador created in Hong Kong. She has been programmed to move like a human, show facial expressions, draw, and even sing.
- Robots – In general, there are a lot of robots around us. They can be used to make deliveries, assist with surgery, serve food and drinks, and perform routine tasks such as cleaning, crop dusting, and sort library books.
- Computerized People – This may sound very Sci-Fi, but people are using radio frequency identification tags (RFID) to “computerize” themselves. This means a small tag, some the size of a grain of rice, is implanted in a person so they can electronically perform tasks with a quick scan. An RFID may be used by people to unlock doors, track location, or even identify status in a controlled environment.
As intelligent as a machine may seem, remember that at the start of it all, it was programmed by a human.
