With artificial intelligence (AI) so sophisticated it can pass the Turing Test, loaded with 175 billion language parameters, middle managers in major manufacturing organizations may be looking over their shoulder. If AI can tell employees what to do, why do they need middle managers? Even hosting a meeting and putting out tea could be made redundant.
Pope Francis may have a more hopeful and probably accurate idea of how AI will contribute to the foreseeable future though—it will serve humanity, even saving us from ourselves. Humans understand context, can build relationships with other people, and connect dots between unrelated possibilities better than machines can. But machines—AI—is fantastic at a number of things that humans are predictably bad at.
And it is here that AI will come to market first—not to replace us, but to cover our back sides when we mess up.
People are unpredictable
One thing that AI will excel at is processes that require more consistency and attention to detail than humans are capable of. All of us can remember times human error had high cost—something as simple as entering a purchase order in Yen instead of British pounds, which would snarl cash management in ways that may be difficult to track down. AI will spot things that are incongruent because it is not thinking about having lunch, taking a coffee break, being yelled at by the boss or considering the next task to be completed.
Matching invoices to purchase orders is also typically not on anyone’s list of favorite things to do, and AI can take on this work and do a better, faster job than people who frankly would rather be performing some other function that is more meaningful for them and uses them for their unique abilities and aptitudes.
At one point in my career, I was sent to a company to unravel why they had automatically matched only 7 percent of their purchase invoices to Goods receipts and purchase orders versus 97 percent which was common in the rest of the company. It turns out that users had configured the system to match on purchase order and advice note. Unfortunately, the goods inwards clerk was never told so he entered any old number in the advice note box so nothing matched.
It took some doing but in a short space of time, more than 4,700 goods received notes (GRNs) were matched to purchase orders. This eliminated serious problems including payments generated automatically to suppliers that had already paid and allocating GRNs to the incorrect purchase orders and projects. AI can match received goods and payments to purchase orders faster and more reliably, and simple but impactful steps like this represent the low hanging fruit for manufacturers’ digital transformation efforts.
Take us to mars
One person who understands AI better than most is Garry Kasparov, who famously was the first world chess champion to lose to AI. He has since championed a collaborative approach to play, where people can contribute their X factor and AI contributes raw computational power to analyze more potential moves than the human ever could.
Kasparov told Jordan Teicher of Medium that we should yield to AI in these areas of superiority and focus on our own unique natures.
“With so much power now brought by machines, we have to find a refuge in our humanity,” Kasparov said. “It’s about our creativity, our intuition, our human qualities that machines will always lack. So, we have to define the territory where machines should concentrate their efforts. This is a new form of collaboration where we recognize what we’re good at and not interfere with machines where they’re superior—even if it hurts our pride.”
In his book, Deep Thinking, Kasparov paints an optimistic picture of this AI-enabled future.
“No matter how good our algorithms are, we can always find better and more creative and ambitious ways to use them,” Kasparov writes. “Why make geniuses dig ditches when they could take us to Mars?”
Read more blogs from Colin Elkins.
Do you have questions or comments?
We’d love to hear them so please leave us a message below.