If I was to ask you to explain the difference between quantum computing, machine learning, and artificial intelligence (AI), in 50 words or less, would you be able to?
No? You’re not the only one.
According to Gartner, AI has increased by 270% in business in the past 4 years. Despite this extreme growth, people are still considerably confused about what AI actually is. AI washing — a term used by vendors to attach themselves to AI when they don’t have a true AI solution — has become the rage, similar to what cloud washing was, 5 years ago.
How to define AI?
AI can be defined as applying logic-based techniques and advanced analysis, (that includes machine learning) to support and automate decisions, take action, and interpret events. AI is a technology that replicates human performance, by typically, learning from it.
So what must be included for a technology to be really AI?
What Really Qualifies a Solution, Product, or Service as AI?
The definition of AI has changed over the 10 years. At the beginning of AI and data science in the 1950s, AI experts were looking to solve general problems. Today, AI is mostly focused on specific problems in specific domains. In order for AI to go in the future, it needs to move from pattern matching and finding the right solution from past experience to solving challenging and difficult cases that require creativity. Despite it, with this evolution has come truths regarding what AI is, and isn’t, to data scientists:
-Ability to make novel discoveries. An AI engine could identify a cluster in high dimensional data that is not easy to be discovered or seen. Retail data or medical Image processing are good examples of it, finding hidden clusters in high dimensional data.
Need to create generalizable rules for the future. It is not AI if a system learns from the past with high accuracy but is not able to generalize to the future. Detection of trends is a good example of work that does not fit in an AI world. While trend detection creates value, an AI system should be proficiently analyzing millions of what-if scenarios to answer and react in the future.
Capability to transform data. AI systems should be able to transform data to be able to learn, infer and analyze. For example, a true AI system might transfer unstructured data (text, voice, etc.) to structured data (categories, words, numbers, etc. ).
Ability to adapt and learn. AI needs to learn from past behaviors. For instance, we expect an intelligent word processing system to start learning our typos, mistakes, acronyms and sentence structure we use as well as our own personal style and our preferred word. We also expect a smart engine to start suggesting more sophisticated words if we are writing an article while suggesting more technical words when we write a product general document. In addition, we expect the same AI system to identify who is using a computer after the first 30 seconds of user activities.
Ability to make decisions in unfamiliar situations. Applying learnings from the past to the future is only the first step for AI. Intelligent systems should be able to infer from old data to not only predict an upcoming event but to attempt to come up with an answer for a new situation that has not been encountered in the past. For example, an autonomous car in Los Angeles that has learned to drive in good weather and heavy traffic, must now learn how to react on an early rainy morning when there is light traffic. Does it drive fast? Slow? What happens if a child crosses the road chasing a ball? A true AI system is able to adjust to the unfamiliar and avoid not only avoiding the child but to expect a child to be behind the next rolling ball.
Ability to work with big data. In our modern age, AI is only as good as the data at its core. Bad data drives bad results. As humans, we process millions of data points in a single day from an early age. AI should move toward processing unlimited amounts of data and be able to learn from every single piece of it.
Ability to remember and forget. One of the main factors of AI systems is the ability to forget knowledge that is not useful. Knowledge that does not have a purpose should be purged. The entertainment industry learned over time that recommendation systems have to consider and learn that people’s interest will shift over time and that you cannot recommend a movie over and over for five years. This simple factor shows the value of time in all AI processing module.
Capability to interpret, describe and prescribe. As much as people are interested in AI magic, they are eager to explain the outcome. AI systems with the ability to interpret and explain can take the human-computer interaction to the next level. Imagine an AI system which can suggest everyone turn off their AC when they are not home to save energy Based on available data, it learns that if the AC should shut off at certain times of the day, turn on at others, shut off when it is a certain temperature outside, or blast cold air when the summer heatwave hits.
Even if the term AI has become more and more popularized and used for organizations promising to leverage the capability, real, purposeful AI is still a unique capability and differentiator beyond just a marketing enhancement.