What happens if we outsource boring but essential work to AI? Research shows that we forget how one can do it ourselves

In 2009, an Air France plane crashed into the ocean, leaving no survivors. The plane’s autopilot system shut down and the pilots needed to depend on their computer-based assistant unable to correct the situation manually.

In 2015, a bus driver in Europe entered the incorrect destination into his GPS device and happily picked up a gaggle of Belgian tourists a detour of 1,200 kilometers within the incorrect direction.

In 2017, U.S. prosecutors overturned a call to release a young person on parole in a call that was later overturned on appeal abruptly modified their minds because an algorithm classified the defendant as “high risk.”

These are dramatic examples, but they’re not at all isolated. If we outsource cognitive tasks to technology—reminiscent of flying a plane, navigating, or making judgments—research shows that this is feasible lose the flexibility to do these tasks yourself. There’s even a term for our tendency to forget information available through online engines like google: the Google effect.

As latest AI technologies promise to automate increasingly activities, the chance of “skill erosion” grows. Our research shows how it will probably occur – and suggests ways to retain the expertise you wish even once you don’t need it day by day.

Loss of capabilities can cripple a corporation

My research shows that the chance of skills loss is definitely missed. In a recent study, my team and I investigated Loss of qualifications in an accounting firm.

The company had recently stopped using software that automated much of its fixed asset accounting services. However, the accountants were unable to finish the duty without them. Years of overreliance on the software had eroded their expertise, and ultimately they’d to relearn their asset accounting skills.

Although the software was rules-based (it didn’t use machine learning or “AI”), it was “smart” enough to trace depreciation and generate reports for a lot of tax and financial purposes. These are tasks that human accountants found very complex and tedious.

The company only became aware of the talents gap after a customer discovered errors within the accounting team’s manual reports. Since the accounting department didn’t have sufficient expertise, the corporate needed to hire the software provider to repair the errors.

How a lack of competence occurs

We found that a ignorance of the automation-assisted task had resulted in a lack of competence. The old adage “use it or lose it” applies to cognitively intensive work as much as anything.

The accountants had no concerns about outsourcing their considering to the software since it worked almost flawlessly. In other words, they fell “Automation complacency“: the belief that “all the pieces is high-quality” while ignoring potential risks.

This had three important consequences:

1. They lost awareness of what automation was doing

2. They lost the motivation to keep up and update relevant knowledge (e.g. tax law) since the provider and software did it for them

3. Because the software was reliable, they not bothered to envision the outgoing reports for accuracy.

This is the way you maintain your skills

So how do you avoid complacency when using AI and other automated systems? Here are three suggestions:

1. Pay attention to what the system is doing – what inputs are getting used, for what purpose, and what might affect its suggestions

2. Keep your competency up to this point (especially when you are legally accountable for the outcomes)

3. Evaluate the outcomes critically, even when the tip results seem satisfactory.

What would that appear like in practice? Here’s an on a regular basis example: Driving with the assistance of an AI-supported navigation app.

Instead of blindly following the app’s instructions, concentrate to road signs and landmarks and pay attention to what you might be doing, at the same time as the app guides you.

Study the map and suggested route before driving to extend your “domain knowledge,” or understanding of the route’s surroundings. This helps you relate your specific path to the broader area, which is useful when you wander off or want to seek out alternative routes.

When you reach your destination, think in regards to the route the app suggested: was it fast, was it protected, was it nice? If not, consider taking a distinct route next time, even when the app suggests otherwise.

Is AI a mandatory companion?

The accounting firm case also raises a bigger query: Which skills are relevant and price retaining, and which should we leave to automation?

There isn’t any one-size-fits-all answer because job skills change over time, jurisdiction, industry, culture and geographic location. However, it’s an issue we must grapple with as AI takes over activities that were previously considered unautomatable.

Despite the difficulties, the accounting manager in our case study believes that the automated software is amazingly useful. In his view, his team was simply caught off guard by complacency.

In a world focused on efficiency and annual or quarterly goals, firms prefer solutions that improve things within the short term, even in the event that they have negative negative effects in the long run. This is what happened within the accounting case: efficiency gains overshadowed abstract concerns about expertise until problems arose.

This doesn’t mean we must always avoid AI. Companies cannot afford to miss the opportunities this brings. However, it’s best to also pay attention to the chance of a lack of competence.