Artificial intelligence generally falls into two broad categories:
Narrow AI: Sometimes referred to as “weak artificial intelligence”, this type of AI works in a limited context and is a simulation of human intelligence. Narrow AI is often focused on performing a single task very well, and while these machines may appear intelligent, they operate under far more limitations and limitations than simpler human intelligence.
General Artificial Intelligence (AGI) – Sometimes referred to as “strong artificial intelligence,” this is the kind of artificial intelligence we see in movies, like robots from Westworld or data from Star Trek: The Next Generation. Artificial General Intelligence is a machine with general intelligence and, like a human, you can apply this intelligence to solve any problem.
Examples of artificial intelligence
Smart assistants (like Siri and Alexa)
Disease prediction and mapping tools
Manufacture of unmanned robots
Improved and personalized healthcare treatment recommendations
Chat robots for marketing and customer service
Robo Stock Trading Advisors
Social media monitoring tools for dangerous content or fake news
Recommendations for TV shows or music from Spotify and Netflix
Narrowness of artificial intelligence
Narrow AI surrounds us and is arguably the most successful AI research to date. Focused on accomplishing specific tasks, narrow AI has seen many advancements in the last decade that have had “significant social benefits and contributed to the nation’s economic vitality,” according to “Preparing for the Future of Artificial Intelligence,” an Obama administration 2016 report.
Some examples of narrow AI include:
Image recognition program
Siri, Alexa, and other personal assistants
Machine learning and deep learning
Much of narrow AI is supported by advancements in machine learning and deep learning. Understanding the difference between artificial intelligence, machine learning, and deep learning can be confusing. Frank Chen’s Venture Capital provides a good overview of how to distinguish the two, noting:
“Artificial intelligence is a set of algorithms and intelligence to try to simulate human intelligence. Machine learning is one of them, and deep learning is a machine learning technique.”
Simply put, machine learning feeds computer data and uses statistical techniques to help you “learn” how to gradually improve on a task, without being specifically programmed for the task, eliminating the need for millions of lines of written code. Machine learning consists of supervised learning (using classified data sets) and unsupervised learning (using unlabeled data sets).
Deep learning is a type of machine learning that manages input through a biologically inspired neural network architecture. Neural networks contain a series of hidden layers through which data is processed, allowing the machine to “dig deep” in its learning, make connections, and weight inputs for better results.
Artificial general intelligence
Creating a machine with human-level intelligence that can be applied to any task is the Holy Grail for many AI researchers, but the pursuit of general AI has been fraught with difficulties.
The search for a “universal algorithm for learning and working in any environment” (Russell and Norvig 27) is not new, but time has not eased the difficulty of creating a machine with a full set of cognitive abilities.
General artificial intelligence has always been an inspiration for dystopian science fiction, with super-intelligent robots rampaging through humanity, but experts agree that it’s not something we need to worry about anytime soon.