During the Covid-19 pandemic when our online engagements had skyrocketed and we spent most of our time online, my friends and I started noticing peculiarities in how the internet presented itself to us. The content that we came across was often personalized to an eerie precision prompting us to joke about how there was a little man, an agent, behind our screens, monitoring our activity so as to be better able to deliver both personalized suggestions and personal attacks. We lovingly named our stalker Gary. As time went by we started referring to Gary as we would an annoying acquaintance who says things that are true but we do not like, and we laughed about how off or on-point he was about certain content he “chose to show us” on the internet. And then we found out that Gary was real. We all have a Gary. It is true, your Gary is an algorithm.
Commonly known as recommendation algorithms or recommender systems, these pieces of code are things of wonder. They learn from your activity on the internet to provide you with content that would best suit your needs and desires, and make sure you do not come across content that may upset or distress you, or that you may dislike. Recommender systems are the foundational entity the “consumer-obsessed” conglomerates of today are based on. They have data on you that you do not even know exists, about every move, every tap, every scroll, and every swipe. It will not be very far-fetched to say that often, if you have been on the internet for long enough, which most of us have, these recommender systems know you better than you know yourself.
But there are limits to how great it is that users are being thoroughly bombarded with the same sort of content throughout their stay on the internet. The “consumer-centric” models of the internet may seem magical and otherworldly but it is necessary to remind yourself that, at the end of the day, you are not just a consumer.
Consumerism.
Just bought a washing machine? How about a year’s supply of washing powder to go with it? While you are at it also look at this clothesline on which you can hang your laundry out and no one will complain about it being dirty or clean because the clothesline looks so pretty. And while we have you here, how about a second Squirtle-themed washing machine since we know that you like Pokemon from your activity on a completely different application?
Targeted advertisements are often spot-on, and they are often not, but they are quite effective at swaying you into believing that you need something that you do not. All activity on the internet is reduced to empirical data points so that it is easier to find out what you are most likely to spend money on. You see a product enough times and you start believing that you need it because it is everywhere on your feed and it is hard to escape. Adomavicius et al., in a study, find that consumer preferences are quite malleable and that they can be constructed. Targeted advertisements are also how most of the sites you log on to generate their revenue, and we are all just a bunch of suggestible people on the internet hoping to be told what to spend our money on. However, you deserve to exist on the internet without being treated in a manipulative way that seeks to extort profit from you.
Parasociality.
Do you feel a special sort of kinship with your favorite internet celebrity who has no idea you exist? Do you find yourself obsessing over every aspect of their internet presence and assuming you know everything there is to know about them from the persona of themselves that they present on the internet? When they do not turn out to be like that constructed image of themselves, do you feel upset, disappointed, betrayed? Like you have been personally robbed of something? Like they owe you an adherence to this constructed image?
Even if that does not describe you it still does a large section of the people surfing the internet. Recommendation algorithms entail constant exposure to certain social media entities and this has resulted in several parasocial relationships that often end in vitriolic Twitter threads and cyber harrassment. Comedian John Mulaney was under the heat in 2021 when it was found that he struggled with substance abuse and had separated from his wife, and his gently crafted on-stage persona of an unproblematic person in a perfect married relationship was challenged. His personal life was the subject of discussion on many internet forums and people felt like they were owed an explanation.
This constant onslaught of news and content regarding these celebrities that recommendation algorithms entail only aggravates the issue of obsessive parasocial relationships which are toxic for both parties involved.
Polarization.
What these recommendation algorithms choose to show you is often contingent upon what is most likely to hold your attention for a few extra seconds. And hence, if it knows what your political ideology is, it will only show you content that you are most likely to agree with. It is important to note here that social media and fake news played a huge part in the results of the 2016 US presidential election, and thus the internet holds power over an entire country’s future. All content is ideologically loaded and it is hard to find anything that does not inform public opinion one way or the other.
Entities that monopolize your attention are often ideologically homogenized and can tend to polarize your views even further. If you like a couple of posts that align with a certain political ideology, chances are you will be led to an entirely polarized corner of the internet from where you are blind to everything else that is not shown to you. The algorithm does not care about which information rabbit hole you are in as long as you are in one. It may not be the site architecture that is entirely responsible for this but it can be changed to combat these issues.
To reiterate the oft-employed adage from The Social Dilemma, “If you are not paying for the product, you are the product.” Your attention is currency and media platforms are vying for it. And these media platforms have only one goal - to keep you on the platform for as long as possible to generate more revenue from advertisers. In the end, these algorithms may seem benign, and they might seem like they are put in place for your benefit, and they are often quite miraculous, but they are also quite powerful and can have consequences if not reined in. In the end, the little man behind our screens may be the punchline of all our jokes, but he controls everything from our perceptions to our practices, in ways we may not fully understand.
Squirtle-themed washing machine 😂