Skip to main content

It's Time to Talk About Algorithms

Gina Rich 2026 10th Grade Lexile: 1160

In an episode of the comedy show “The Office,” hapless business manager Michael Scott is driving a car and using GPS for directions. GPS navigation systems use algorithms, which are a set of instructions for performing a task. Michael trusts the technology so much that when the system tells him to turn right, he does exactly that — and drives into a lake.

From driving directions to the personalized “For You” page on our social media apps, algorithms are everywhere. But technology doesn’t always get it right. A faulty algorithm could give us bad directions, disqualify us from our dream job, or even land us in jail. One high-profile case occurred in 2020, when a man named Robert Williams was arrested for a crime he didn’t commit — all because a facial recognition algorithm incorrectly identified him.

Algorithms are meant to make our lives easier. How, then, do they go so wrong? And despite the negative publicity they’ve received, are algorithms still worth using?

In Algorithms We Trust?

One problem with algorithms is that we tend to assume — as Michael did — that the technology can’t make mistakes. “Humans often blindly follow whatever technical devices tell us because many of us believe technology is superior and more objective,” Hilke Schellmann, a journalism professor at New York University, writes in her book The Algorithm: How AI Decides Who Gets Hired, Monitored, Promoted & Fired & Why We Need to Fight Back Now.

Perhaps because we put so much trust in algorithms, we fail to get curious about how they work or why they make a particular recommendation. In fact, often “the developers who build the tools don't know how they function. That is very problematic,” says Schellmann.

And when it comes to social media, algorithms have grown so adept at serving up content we like that we tend to think less critically about what we’re consuming. In a 2023 article for The New York Times, U.S. Senator Chris Murphy argues that algorithms have led to “the death of exploration, trial and error and discovery.” Kids are “not learning how to be curious, critical adults,” he writes. Senator Murphy and other lawmakers introduced the Protecting Kids on Social Media Act, which aims to prevent social media companies from delivering algorithm-generated content to users under 18.

But convincing kids that the customized feed on their favorite apps is bad for them may be difficult. Many young users rely on the algorithms to deliver entertaining content. 15-year-old Greta* likes that her TikTok “For You” page is different from those of her friends because it feels personalized. The content is so compelling she often loses track of time. “Because it’s stuff that I like, I watch it the whole way — I don’t just scroll,” she says. Her experience suggests how effective algorithms have become in keeping users on platforms longer.

The Problem of Bias

Perhaps there’s nothing inherently wrong with being shown content we enjoy, whether that’s cute puppy videos or the latest popular salad recipe. But many algorithms are riddled with bias that can impact which voices are amplified and which are silenced. On TikTok, for instance, certain videos tend to go viral, yet exactly how the algorithm makes that happen remains a mystery to content creators.

This bias can be frustrating for both users and creators on social media. In an article in Forbes, Annie Brown, founder of the inclusive creative sharing platform Lips, explains how algorithmic bias against certain groups results in a phenomenon referred to as shadowbanning. “Apps can be programmed via algorithms to avoid and even eliminate specified content, effectively ‘pushing down’ users’ posts to the point where their content is hidden from other users in the community,” she writes.

But how do algorithms become biased in the first place? Algorithms respond to and learn from users’ actions. In this way, human bias makes its way into the algorithm. If you’re less likely to interact with a certain type of content, the app will show you less of it. However, many algorithms start off biased. According to Brown, “The unconscious biases of the developers are embedded in the systems they create.” And once an algorithm is created, there are more opportunities to introduce bias. “Algorithms are trained by data gathered from from human history, a history replete with violence, inequity, bias and cruelty,” Brown explains.

So while users of an app may be less likely to interact with content based on their own biases, some may never get a chance to decide for themselves because the algorithm has already decided for them. And for creators on these platforms, it means they could spend hours on a post or video only for no one to ever see it. Algorithms can also impact what content gets created in the first place. In an article published in the MIT Technology Review, senior editor A.W. Ohlheiser explains how creators have fundamentally changed the content they create. “Some creators adapt their entire approach to producing and promoting content in response to the algorithmic and moderation biases they encounter.”

Aiding – Not Replacing – Human Decision-Making

Though algorithms are far from perfect, experts believe the technology has advantages as long as we design and use it appropriately.

One optimist is Tomas Chamorro-Premuzic, a psychology professor and author of the book I, Human: AI, Automation, and the Quest to Reclaim What Makes Us Unique. In an article for Fast Company, Chamorro-Premuzic writes that while algorithms can be biased, we can teach them to “unlearn” these biases. By contrast, a human can never unlearn or ignore “that the person in front of them appears to be female, male, white, Black, rich, poor, attractive, unattractive, old, young, etc.”

Chamorro-Premuzic views algorithms as a helpful tool that can free us up to focus on “humane tasks,” like connecting with people emotionally.

In Schellmann’s view, algorithms can be useful for “low stakes'' problems, such as blocking spam emails from our inboxes. But for decisions that could truly change a life — whether to hire someone, approve them for a loan, or pay for their medical care — “we really have to think through it,” says Schellmann. “Is this a fair decision? How do we come to this decision?”

As algorithms grow more sophisticated, asking these questions will be critical, says Schellmann. “Now’s the time to talk about these tools and intervene.”

*Name has been changed