“While only a few years ago the only people who would utter the term were Ray Kurzweil and the few other courageous zealots from the ‘old days’ who couldn’t give up on the vision,” says Reynolds, “now we see it discussed on the local TV news, on the covers of general audience magazines, and, helpfully, in the ‘failing’ New York Times.”
And already, we’re living in a world surrounded by AI. “When you look around at what areas of life are now incompletely serviced by intelligence that could be delivered by machines, you can get a sense of AI cropping up all over the place,” he continues, “from the self-stocking refrigerator to the interesting bot companion who converses daily with your aging grandparents.”
However, how much does the average person know about AI? While it seems to be constantly discussed, there’s little added to the topic on expected developments and other details.
Reynolds says, “As much attention as the progress and the promise of technology has gotten in the media, the resulting cloud of hype has created more confusion than understanding around the terms that it uses, like ‘machine learning’ and ‘deep learning,’ and most especially around ‘AI’ versus ‘cognitive computing.’”
AI in business
One important aspect to keep in mind is how businesses and individuals regard their reliance upon AI.
Reynolds says, “In a big data environment, because of the sheer volume and complexity of the interactions among systems, human analysts are increasingly dependent on machines to do the preparation and the ‘reading’ of their own reports, the analyses of their emerging trends, and, in the most advanced of today’s systems, even the process of coming up with recommendations for human decision-makers to turn into actions.”
He continues, “For any business, the first question to ask about any AI initiative is, ‘Do we have an AI problem?’ Is the data truly huge and unmanageable? Is it critical to know in real time or near real time what is going on in the business? Do our people need to have their hands on the pulse of the full matrix of data about a particular customer relationship?”
One of the toughest aspects to come to terms with regarding AI is that the computer systems we’ve become so reliant upon will continue to change and they may at times come to unexplainable conclusions.
“AI systems are probabilistic at the core, and they don’t behave in the way that the deterministic systems that we are used to do,” Reynolds says. “This is all enough to give a chief compliance officer permanent migraines.”
While we’re not quite at the point of living alongside sentient robots, there are ways that AI can be brought into a job to make it more efficient.
Trial and error
“To the extent that professional vocations now include repetitive decisions within a limited range of options, much of what professionals now do could be a candidate for automating with an AI-based system. The machines are not about advanced human reasoning yet, however, so more complex and collaborative intellectual work is not now addressable with AI.”
“Progress will be rocky, the tempo will be uneven, mistakes will be made – perhaps development will stop for a while as reassessments take place and adjustments are invented.” But despite the hype, Reynolds says, “The groundwork being laid now will continue to support a new class of intelligent systems.”