|
The definition for Artificial Intelligence (or AI) is simple. We’re not talking about your little brother when he makes up his answers on his math quiz (he’s got real intelligence, that’s just called “not studying”). Artificial Intelligence is intelligence exhibited by machines. While what exactly “intelligence exhibited by machines” means is often in the eye of the beholder, generally speaking, AI is accepted by most to be referring to when a machine or computer makes independent decisions based on what it encounters in a given situation-- and not based on a set of programming.
There are many different kinds of AI that computer scientists are working on. A few of the more prominent are: Reasoning; Planning; Learning; and Perception. Scientists and programmers are trying to develop computer software that can learn and reinforce wanted outcomes (called Goals) and to do so without being explicitly programmed prior for a given situation. Currently there are two different approaches to creating AI: Hard (or Symbolic); and Soft (or Sub-Symbolic). Symbolic means the data fed to the AI entity is in the same format as data provided to humans, while sub-symbolic means that the data is instead fed to the AI in a form more similar to a computer search, translating the data into what’s called Assembly Language. Sometimes it can be difficult to tell what is and isn’t AI. For instance, if a computer is programmed with a sufficient level of statistical ability, it might be able to make more accurate predictions about what will happen than a person can, but that doesn’t necessarily mean the computer is smarter-- it’s just able to work through complex equations much more quickly to formulate a prediction. Many stockbrokers now use very complex computer programs to assist in the selection of stocks to be purchased, but again, it doesn’t mean these programs are smarter-- they’re just tools offering support to the stockbroker who will make the ultimate decision based on many factors. The state of the science of AI has taken giant leaps in the past few years, as the amount of computing power has grown significantly. The more computing power an AI entity has, the more likely a true, sentient AI can be developed. With that said, true, sentient AI-- though likely-- is not necessarily a given. A famous computer scientist, Alan Turing, for instance, thought it was impossible for a computer to ever think as humans think-- but also conceded that if, when questioned, a computer’s answer could not be differentiated from a human response, then the computer is as intelligent as a human. This test (the Turing Test) is used often today to help gauge what level an AI has currently reached. Early Artificial Intelligence is already here, and odds are you already interact with it to some degree in your everyday life. Self driving automobiles, planes, and even ships all use AI; a form of AI exists in most smart phones and smart home devices, called Virtual (or Digital) Assistants; AI can even be found in a number of smart appliances, like refrigerators and vacuums! Some scientists are afraid that AI might get out of hand and that computers will learn to do things not beneficial to us humans. Movies are ripe with this kind of anxiety. Some fear that if computers do ever attain a level of intelligence that is superior to humans, they might combine their superior cognitive abilities to become masters of the world. Other scientists view such fears as nonsense and believe that these scientists are forcing human logic onto a computer “brain” that would have no use for it. Only time will tell who’s right. Should we worry about the computers taking over? Or is this just something that happens in post-apocalyptic movies? Well, we won’t know until we know-- but if it’s keeping you up at night, remember-- a computer is ultimately just another machine, and we’ve always managed to keep our machines under control. For now. |
|
GLOSSARY
Artificial Intelligence (AI) the theory, development and implementation of computer systems able to perform tasks that normally require human intelligence, such as visual perception, speech recognition, decision-making, translation between languages, etc.
Assembly Language a low-level code of symbols converted by an assembler.
Sub-Symbolic (Artificial Intelligence) methods of sub-symbolic artificial intelligence manage to approach intelligence without specific representations of knowledge, unlike symbolic methods.
Symbolic (Artificial Intelligence) symbolic artificial intelligence is the collection of all methods in artificial intelligence research that are based on high-level “symbolic” (human-readable) representations of problems, and was the dominant paradigm of AI research from the mid-1950s until the late 1980s.
Turing Test the Turing test, developed by Alan Turing in 1950, is a test of a machine’s ability to exhibit intelligent behavior equivalent to, or at least indistinguishable from, that of a human.
Virtual (or Digital) Assistants a digital assistant is a software agent that can perform tasks or services for an individual.
Assembly Language a low-level code of symbols converted by an assembler.
Sub-Symbolic (Artificial Intelligence) methods of sub-symbolic artificial intelligence manage to approach intelligence without specific representations of knowledge, unlike symbolic methods.
Symbolic (Artificial Intelligence) symbolic artificial intelligence is the collection of all methods in artificial intelligence research that are based on high-level “symbolic” (human-readable) representations of problems, and was the dominant paradigm of AI research from the mid-1950s until the late 1980s.
Turing Test the Turing test, developed by Alan Turing in 1950, is a test of a machine’s ability to exhibit intelligent behavior equivalent to, or at least indistinguishable from, that of a human.
Virtual (or Digital) Assistants a digital assistant is a software agent that can perform tasks or services for an individual.
