We bandy about the term “artificial intelligence,” evoking ideas of creative machines anticipating our every whim, though the reality is more banal: “For the foreseeable future, computers will not be able to match humans in their ability to reason abstractly about real-world situations.” This is from Michael I. Jordan, one of the foremost authorities on AI and machine learning, who wants us to get real about AI.
SEE: Robotics in the enterprise (ZDNet/TechRepublic special feature) | Download the free PDF version (TechRepublic)
“People are getting confused about the meaning of AI in discussions of technology trends—that there is some kind of intelligent thought in computers that is responsible for the progress and which is competing with humans. We don’t have that, but people are talking as if we do,” he noted in the IEEE Spectrum article.
Instead, he wrote in an article for Harvard Data Science Review, we should be talking about ML and its possibilities to augment, not replace, human cognition. Jordan calls this “Intelligence Augmentation,” and uses examples like search engines to showcase the possibilities for assisting humans with creative thought.
And, to be clear, machines are much better at some things. For instance, people could do low-level pattern-matching but at a significant cost, whereas machines are able to perform such mundane tasks at relatively little cost. Another example is that ML is broadly used for fraud detection in financial services. We could have people poring over millions upon billions of transactions, but it makes more sense to point computers at the problem.
We know that most AI projects fail. In Jordan’s emphasis on ML over AI, there’s perhaps a clue as to why AI projects fail (inflated expectations) and how to make ML projects succeed (tightly define projects to augment, not supplant, human actors).
SEE: Artificial intelligence ethics policy (TechRepublic Premium)
The more we get “real” with AI, in other words, the more likely we’ll find success. Fortunately, Jordan wrote, most of the time when we’re talking about AI, we really mean ML. “ML is an algorithmic field that blends ideas from statistics, computer science and many other disciplines to design algorithms that process data, make predictions and help make decisions,” he wrote in the Harvard Data Science Review. ML is essential to “any company in which decisions could be tied to large-scale data,” he added.
So…the first rule for success in AI is to stop doing AI, and instead consider data science problems as fundamentally about ML, about finding patterns in large quantities of data. It’s not Jetsons, but it’s real.
Disclosure: I work for AWS, but the views expressed herein are mine.