ATHENS, Ga. — Do you find yourself reaching for the calculator, even for the really simple math problems? There’s a lot of concern these days that technology, like artificial intelligence, is too smart for its own good. Despite fear over how intrusive these algorithms are becoming, a new study finds people are actually more willing to trust a computer than their fellow man.

Researchers at the University of Georgia say this is especially true when people find tasks too challenging to handle alone. However, it’s not just the “heavy lifting” humans are running to computers for help with. From choosing the next song in the playlist to finding better fitting pants, algorithms are making more and more of the daily decisions in people’s lives — whether they realize it or not.

“Algorithms are able to do a huge number of tasks, and the number of tasks that they are able to do is expanding practically every day,” says Eric Bogert, a Ph.D. student in the Terry College of Business Department of Management Information Systems, in a university release. “It seems like there’s a bias towards leaning more heavily on algorithms as a task gets harder and that effect is stronger than the bias towards relying on advice from other people.”

Letting the computer do the work

Researchers evaluated the responses of 1,500 individuals tasked with counting the people in a series of photographs. The team also supplied participants with suggestions on how to do this, generated either by other people or computer algorithms.

As the crowd in the photos got bigger and more difficult to count, volunteers were more likely to turn to the computer’s suggestions rather than go with their own gut or the “wisdom of the crowd.”

Study co-author Aaron Schecter says counting tasks are a perfect way to measure trust in computers. As the number of people grows in each photo, the task becomes objectively harder for humans to perform. The UGA team adds these tasks are also the kind of problems humans expect computers to be good at solving.

“This is a task that people perceive that a computer will be good at, even though it might be more subject to bias than counting objects,” Schecter explains. “One of the common problems with AI is when it is used for awarding credit or approving someone for loans. While that is a subjective decision, there are a lot of numbers in there — like income and credit score — so people feel like this is a good job for an algorithm. But we know that dependence leads to discriminatory practices in many cases because of social factors that aren’t considered.”

Replacing human biases with computer biases?

Just because a computer program is merely a pile of data doesn’t mean biases don’t exist. Schecter notes facial recognition and hiring algorithms have both been criticized recently over cultural biases built into their programs. These can lead to inaccuracies when matching faces to identities or screening qualified job candidates.

Although simple counting tasks won’t display bias, researchers caution it’s important to know how machines arrive at more complex decisions. This study is also part of Schecter’s larger report on human-machine collaboration, funded by the U.S. Army.

“The eventual goal is to look at groups of humans and machines making decisions and find how we can get them to trust each other and how that changes their behavior,” Schecter concludes. “Because there’s very little research in that setting, we’re starting with the fundamentals.”

The team is now studying how people rely on algorithms to make creative and even moral judgments. These traditionally human tasks include writing original works of literature or setting bail for criminals.

The study appears in the journal Scientific Reports.

About Chris Melore

Chris Melore has been a writer, researcher, editor, and producer in the New York-area since 2006. He won a local Emmy award for his work in sports television in 2011.

Our Editorial Process

StudyFinds publishes digestible, agenda-free, transparent research summaries that are intended to inform the reader as well as stir civil, educated debate. We do not agree nor disagree with any of the studies we post, rather, we encourage our readers to debate the veracity of the findings themselves. All articles published on StudyFinds are vetted by our editors prior to publication and include links back to the source or corresponding journal article, if possible.

Our Editorial Team

Steve Fink


Chris Melore


Sophia Naughton

Associate Editor