What Are Algorithms, Really? The Truth Behind the Technology Shaping Our World
Written by Massa Medi
They're everywhere. Algorithms lurk inside the social media apps we scroll through countless times a day, power our search engines, and even decide who might be our soulmate on dating platforms. But for all this talk about algorithms, one big question remains: what are they, really? Are they something to be worried about? Do they think for themselves, or is that just science fiction?
Like any researcher—or, let's be honest, anyone confused and curious in 2025—I started where every great investigation begins: the search engine. Typing in “algorithm,” I was greeted with the most official-sounding definition I could find: A process or set of rules to be followed in calculations or other problem-solving operations, especially by a computer.
That sounds almost like a riddle. After an hour falling down the internet rabbit hole, reading definition after definition, everything started to blur together. What do algorithms actually do? Determined to get some real answers, I reached out to a bona fide expert—Professor Victoria Nash from the Oxford Internet Institute.
Algorithms, Explained (with Cake!)
Professor Nash broke it down in the simplest way possible: think of an algorithm as a set of instructions—just like a recipe. For example, a recipe tells you exactly how to combine flour, sugar, eggs, and a dash of hope to create a cake. Replace flour and eggs with data, and you've got a computer algorithm: it’s a set of instructions telling a program how to process information and what result to produce at the end.
Hearing that made algorithms seem less mysterious and a whole lot less threatening. But still, the same process that gets me a cake somehow also steers my search results. How could such a simple concept drive such powerful technology? Enter Dr. Bernie Hogan, also from Oxford, who invited me for a behind-the-scenes look.
A Noisy Tour Behind the Scenes
Dr. Hogan met me in the university's sprawling data center. Picture this: endless rows of hot, humming server racks, all packed together in a high-security, nondescript room that almost nobody ever sees, but on which modern life seems to depend. The noise is constant—a kind of mechanical white noise you can feel in your chest.
“Each one of these belongs to a different department,” Dr. Hogan explained, gesturing to the blinking machines. “They're doing different kinds of calculations. So there are thousands of algorithms running inside these boxes. Actually, scratch that: billions.”
As we retreated to a quieter space to talk details, Dr. Hogan emphasized why we need algorithms to begin with: it's all about data. We have mountains of it—lists of towns, endless tweets, flight prices, you name it. To make sense of this avalanche of information, algorithms step in, following their instructions to sift, sort, and stack the data.
“Algorithms calculate, based on a bunch of features, the things that will put something at the top of the list, and something else at the bottom,” Bernie said. For example, when you search for a flight online, the order of prices you see may change depending on the day, your location, or even the device you use. That’s an algorithm in action—a sometimes mysterious, not-always-transparent one.
The Black Box Problem
If it’s all just simple instructions, should we be scared? Dr. Hogan raised the concept of the “black box.” Essentially, we don’t always know exactly what an algorithm’s instructions are, especially when they belong to a private company. This can lead to real-world consequences, like when flight prices change dramatically based on mysterious factors, and we have no way of knowing why.
But what about so-called “thinking” algorithms? Are we approaching the age of self-aware computers? “We wouldn’t say algorithms think,” Dr. Hogan replied, “but they can learn. Algorithms can learn from other algorithms and even generate their own instructions.” At the end of the day, though, the foundation is always the same: data goes in, millions of instructions are followed, and a result comes out.
What Does an Algorithm Look Like?
Frustrated that I’d never seen an algorithm in the wild, I paid a visit to one of the UK’s top coding schools. On arrival, a buzzing classroom of aspiring coders greeted me, all hard at work, blocks of code filling screens.
“So this is code, right? But what’s the difference between code and an algorithm?” I asked.
An instructor broke it down: coding is how you write an algorithm in a way a computer can understand. In this class, they were using Scratch, a super-accessible programming language where you drag and drop colorful blocks that represent instructions. For this lesson, they were writing instructions for a drone to perform tricks—loops, flips, turns—at the press of a button.
Battle of Man vs. Machine: The Drone Race
To really put algorithms to the test, we set up a classic human-versus-machine challenge: a drone programmed by Scratch (and its perfectly consistent algorithm) versus…me, last of the analog humans. The course: through a hoop, do a flip, race around, land, and repeat three times.
The premise was simple. The algorithm-powered drone should execute every movement identically, every time, without a hitch. Humans, as you might guess, are somewhat less predictable. Ready, set, go!
I fumbled through the controls with a combination of hope, panic, and random button pressing. The algorithm, meanwhile, glided the drone through each step smoothly, no drama, no deviations. As the race finished, the winner was obvious: the algorithm, cool as a cucumber, completed all three laps perfectly.
“That’s the beauty of algorithms,” the instructor noted. “No human error. The computer will never get distracted, never make a mistake—unless, of course, the person writing the instructions gets it wrong in the first place.”
Should We Be Afraid?
That raises a whole new realm of discussion: ethics and control. If algorithms are making so many decisions in our world, are we handing too much over to them? Will they take all our jobs? Will they erode our human skills, or even “de-skill” us by making us overly dependent?
The consensus: yes, there are risks. Over-reliance on algorithms could dull our own decision-making abilities. Yet, they also bring enormous benefits: efficiency, speed, and the ability to spot patterns that humans often miss. The key is in responsible use—algorithms should be our tools, not our masters.
The Big Picture: Who’s Really in Charge?
Here’s what I’ve learned: despite all their mystique, algorithms are fundamentally simple. Data goes in, the instructions are followed, and an outcome pops out the other end. Across the world today, algorithms are at work in the criminal justice system, social care planning, credit checks, and so much more. These are not just cyber-puppeteers controlling your dating matches or ad feeds; they directly impact human lives.
The crucial question isn’t about algorithms themselves. It’s about who controls them, and where the data they use comes from. With great (algorithmic) power comes great responsibility. Transparency, oversight, and ethical standards all matter now more than ever.
Thanks for reading! If you found this deep dive enlightening, don’t forget to subscribe for more explorations into the technology shaping our world. See you next time!