The media update team takes a look under the "bonnet" of the algorithm world to find out what these pieces of code are, and how they have been changing the way the world works.

Algorithms are trained to solve problems

An algorithm, simply put, is a set of steps for solving problems that run on a computer program. The problems are solved by running through a number of processes, which are defined by certain rules. Although this technology is run on computers, there is still a human element involved at the creation stage. Software developers create algorithms so that intended results comes from particular inputs.

Algorithms can be trained using machine learning and can also be pre-programmed to perform specific duties.

Algorithms, like the machines they run in, are considered to be a technology in their own right. A well-designed algorithm running in an old computer would produce results faster than a poorly designed algorithm, used for the same reason, running in better hardware.

An algorithm dictates what you see after you do a Google search, whose posts appear on your Facebook or Twitter timeline, can decide whether you will be given an interview for a job you applied for, or change your car insurance premium.

Algorithms can change society

A noticeable effect algorithms are having upon society is, through the way political discourse is funnelled to the public.

Will Oremus, senior technology writer at Slate, notes that prior to March 2016, the tweets a person saw on their timeline were, with minor exceptions, from people you followed, in chronological order.

However, after Twitter implemented an “algorithmic timeline”, Oremus argues that it is reinforcing people’s biases and “abetting their construction of alternative realities – not a marketplace of ideas, but a battlefield pocked with foxholes”.

In an article for the Guardian, Alex Hern pointed to a 2015 study, which suggested that over 60% of Facebook users did not realise that their feeds are curated at all. Users believed every story their friends shared and ‘liked’ pages that appeared in their news feed, without asking if they were being served misinformation. 

“In reality, the vast majority of content any given user subscribes to will never appear in front of them. Instead, Facebook shows an algorithmic selection based on a number of factors,” says Hern.

As Forbes staff writer Parmy Olson notes in her piece about how Facebook helped Donald Trump to be elected as president of the United States, the decisions made about what people see on their Facebook news feeds are decided by an algorithm, which surrounds users with “Yes Men”, damaging people’s ability to think rationally.

Algorithms are an important technology that needs human oversight

Automation has transformed agriculture and industry, as Luciano Floridi, Oxford University philosophy and ethics of information professor, writes in the Financial Times. In a society where algorithms and other automated process are “increasingly apparent”, the important question is how far can technology be trusted to make decisions instead of people?

Floridi says that putting humans back into the equation is the best protection against misfires. “Trust depends on delivery, transparency, and accountability,” he says.

“The same holds true for algorithms. We trust them when it is clear what they are designed to deliver, when it is transparent whether or not they are delivering it, and, finally, when someone is accountable – or at least morally responsible, if not legally liable – if things go wrong.”

What is required, he argues, is a system of design, control, transparency, and accountability overseen by humans.

“Algorithms are the new herd. Our future jobs will be in the shepherding industry,” says Floridi.

Want to stay up to date with the latest media news? Subscribe to our newsletter.

Algorithms form part of the artificial intelligence journey. Developing those capabilities in-house can lead to a significant competitive edge for your business. Read more in our article, Bringing innovation in-house is vital in the AI age