You might not realize it, but algorithms are mediating your life constantly, making choices for you, shaping your perceptions and influencing what all of us believe.
Algorithms are constantly working behind the scenes, mysteriously influencing what you watch, read, and listen to. They can determine whether you will get that job you want, if you will get the loan you applied for — even if the traffic light ahead is green or red. In the future they may take your job, save your life, or kill you.
What Is An Algorithm?
Don’t worry if you don’t know much about algorithms — few people do. In its simplest form an algorithm is just a way to reduce problems into a set of rules to which data can be applied to achieve specific outcomes. In other words, an algorithm is a digital recipe made by computer scientists rather than chefs.
Though simple in concept, algorithms are extremely powerful when combined with advanced computing power, vast data sets, machine learning and artificial intelligence.
Who Uses Them
Different groups use algorithms for different reasons, including:
- Publishers use them to create recommendations about the things you should read, listen to and watch
- Search engines use them to surface information that the algorithm thinks you should see
- Healthcare groups use them to optimize patient outcomes and reduce medical errors
- NASA uses them to hunt asteroids
Uber uses algorithms to direct human drivers, just as human dispatchers once did. Uber also uses secret versions of them them to manage surge pricing, just as Airbnb does. Tinder uses a secret algorithm to rate your desirability.
The list could go on and on. Algorithms are pretty much everywhere in our digital lives, lurking behind dating sites, business processes, financial markets, Facebook and plenty more.
Quick Challenge: On average, how many algorithms do you think influence your decisions every day? Scroll down to the bottom to see what others said.
They Live in Unregulated Black Boxes
Algorithms create tremendous advantage for those who deploy them, so they are frequently treated as trade secrets. Despite the profound influence they have on people’s lives, these “black box “ versions lack any sort of transparency that can ensure fairness, accuracy, or accountability.
Mysterious and opaque, oftentimes you can’t completely know how they work, what data they are using, or how accurate they are.
Black box algorithms are problematic for a number of reasons:
- Human bias can be intentionally or inadvertently built in
- Black box algorithms can crash systems and markets
- Data sets used by them may be inaccurate or culled from illegal sources
- Black box algorithms lack the kind of public audit functions that are associated with stable trusted tools of commerce and finance and fair trade
This lack of transparency and accountability has caught the eye of regulators and consumer watchdogs, who worry that a lack of transparency and accountability can lead to abuse. For instance:
- The EU is investigating Google for antitrust violations, accusing the search giant of gaming its search algorithm to artificially promote Google products and services over those of competitors
- The Federal Trade Commission is setting up a new Office of Technology Research and Investigation (OTRI), which among other things will examine issues related to algorithm transparency
- The UN is hiring a global data privacy investigator whose purview may include algorithms
- They can be discriminatory. An Upshot article by The New York Times describes various ways that biases – intentional or not – can be built into an algorithm
An especially relevant example of how algorithms can be misused came to light in 2015, when authorities discovered that Volkswagen coded one into its software for diesel cars in order to hide illegal emissions. Dubbed “dieselgate”, the scandal cost VW at least $25 billion.
Our Future With Algorithms
Clearly algorithms are already a big part of our lives. But soon they will become even more prevalent as rapid advancement in Artificial Intelligence (AI), machine learning and robotics merge with algorithms.
Already developers are working to infuse algorithms with advanced reasoning capabilities and even creativity that closely resemble that of humans. Over time they should be able to automatically improve themselves in a virtual evolutionary cycle.
Put simply, the future of algorithms is at once wonderful and terrifying:
- They will likely help you be see, hear and sense things
- They will enable autonomous swarm air and naval fleets, global financial market flash crashes, and spy agency snooping
- They will use facial recognition to discover people’s feelings and predict emotional reactions
Massive Job Loss Fears
Some believe that they will soon contribute to massive global employment disruptions. In a 2013 study, Oxford researchers predicted that up to 50% of US jobs are vulnerable to computerization in the next 20 years. Some of the jobs that are most at risk include:
- Truck Drivers
- Mail carriers
- Tax preparers, telemarketers, data entry professionals
- Cashiers, clerks, and cooks
Many of these job functions are regulated by the types of rules and processes that algorithms excel at.
Algorithm-driven shifts in employment might create massive economic changes on a global scale. As the economist Paul Krugman put it:
“Smart machines may make higher GDP possible, but they will also reduce the demand for people—including smart people. So we could be looking at a society that grows ever richer, but in which all the gains in wealth accrue to whoever owns the robots. “
The Future Of Mankind
As if jobs weren’t enough to worry about, some leading thinkers believe that artificial intelligence presents a surprisingly substantial risk to the future of mankind. Machine learning and AI algorithms may overtake humans in fundamental ways, say critics like Stephen Hawkings, who postulates that that humans could be superseded by machines .
Some tech luminaries are also worried. Elon Musk, the polymath entrepreneur, worries about AI and algorithms that could inadvertently start a war, according to an interview in Fortune.
In what is perhaps the strangest twist of all, AI machine learning is evolving at such a rapid rate that it may soon lead to the creation of algorithms so advanced that they cannot even be understood by humans. In such a world humans will have to blindly trust that the machines are working in their best interests.
Yes, They Have An Upside Too
To be fair, algorithms have an upside too. When applied to healthcare, human safety, agriculture and other good causes, they can improve lives in many ways.
For instance, financial adviser robo-advisers have a fiduciary responsibility to serve your best interests, whereas some financial professionals do not.
But it may be the dark side of algorithms- market manipulation, unchecked AI, emotional manipulation, dangerous military systems, greedy commercial operators – that we should pay the most attention to.
Challenge Answer: There’s really no way of knowing how many algorithms influence our lives on a daily basis, but it’s a lot. We asked the Voat.com community (the new Reddit competitor), with the most common reply being “too many to count”.
To understand more about algorithms, consider Pedro Domingo’s book “The Master Algorithm“. In the book, Domingo provides an introduction to machine learning and algorithms and delves into the search for a master algorithm that “can derive all knowledge in the world—past, present, and future—from data.”