“In the 21st century, we won’t experience 100 years of acceleration, it will be more like 20,000 years,” writes Ray Kurzweil, who’s predicted the rapid rates of technological evolution that we’re seeing today. Software and computing are at the forefront of what’s being called the Fourth Industrial Revolution, in which we’re moving from automation to machine learning, which grows more sophisticated by the day—with huge implications for businesses across industries.
Artificial Intelligence (AI) is leading the way, and it’s more than just a buzz term. Total business value derived from AI is expected to reach $1.2 trillion in 2018, and rocket to $3.9 trillion by 2022. There’s a clear and immediate impetus for businesses to invest in AI, but the learning curve is quite steep. With that in mind, here are some industry terms you need to know when starting down the path of AI investment. Armed with some basic knowledge, you can start to develop a sense of how AI may factor into the development of your business.
In general, this term refers to a computer system that possesses a human-level intelligence. The Turing test—named for Dr. Alan Turing, who initially proposed it (in 1950!)—has long been considered the benchmark for identifying true artificial intelligence. The test is deceptively simple: If a machine can fool a human interrogator into believing it is human, the machine would be considered truly “intelligent.”
In recent years, “artificial intelligence” has come to refer to the software and algorithms that process information, derive insights, and make decisions based on vast sets of data generated by today’s connected population. These insights and decision-making are at the heart of what makes AI so valuable to the business world.
Machine learning (ML) involves establishing a digital model for a real-world process, then feeding data into that model to build a prediction engine. Once the model has been successfully trained, new predictions can be made. The model can be further refined by comparing these predictions with actual events as they happen, so the model can evolve over time.
Like artificial intelligence, machine learning can also refer to the set of software and algorithms that are used to build these real-world models.
A blanket term that refers to the various applications of machine learning and artificial intelligence in solving real-world problems. Computer vision, speech recognition, and natural language processing are all examples of cognitive computing. Major technology players like Microsoft, Google, and Amazon provide APIs that can integrate the power of the AI/ML into many applications.
A neural network is a specific type of machine learning algorithm, albeit a very powerful one, originally modeled on the way neurons in the human brain pass electrical signals between each other to form cognitive processes. The idea is that by feeding data through multiple interconnected layers of nodes, we can train the entire network to recognize patterns in the data. Neural networks can be used to convert handwriting to digital text or diagnose cancer from medical images.
Deep learning has become a buzzword recently, as there have been some impressive gains in certain problem domains like image classification, speech recognition, and natural language processing. Under the hood, deep learning is implemented by specialized neural networks. In fact, the “deep” in “deep learning” refers to the number of layers in the neural network leveraged by the deep learning model.
Natural Language Processing (NLP)
A marriage of linguistics and AI, the goal is to translate human communication (both oral and written) into a structure that machines can process and respond to intelligently. Think Siri, Alexa, and Cortana. There’s a lot of information encoded in our communications, beyond the words themselves: tone, emotion, intent, etc. Active research in NLP seeks to understand how we can extract these higher-order pieces of information.
Computer vision involves processing visuals via machine learning algorithms in order to extract information contained in those images. Examples include recognizing people, places, and things, or extracting meaningful text from an image (i.e., Optical Character Recognition).
Breaking down the concept of AI unearths its vast potential. Companies are leveraging the technology to more deeply understand people, evolve their own processes, and create better experiences inside and outside of their organizations. Look out for our next post, in which we’ll take a deeper look at emerging trends in AI and how it’s driving business transformation.