The Power of Algorithms

Algorithms and AI play an increasingly influential role in how we experience the world, the decisions we make, and our perception of reality. So how can marketers ensure that they’re using this technology responsibly?

Dr. Renée Richardson Gosline, a senior lecturer and research scientist at MIT Sloan School of Management, explains that marketers should first consider whether an algorithm is necessary to support brand activities and to what degree the algorithm is designed with humans in mind. 

Transcript

I think it’s actually quite difficult to go through the day, any day, without having a machine curate what you see and mediate your experience with the world around you. We’re already at a point where what we choose to eat, who we choose to date, how we drive, and what directions we take are being affected and mediated by technology.

I think what’s important about that though, is that that increase in spread is also being accompanied by a decrease in the consciousness around the fact that this is happening. So it’s become so natural for us to have algorithms and artificial intelligence curating our world. And for us to seek information through these means that we kind of can go on autopilot – it’s so comfortable. So I think as we look to the future, the question becomes not just spread, it becomes whether we even notice it at all.

This filtering and this ability to see the world through the eyes of what the machine has decided is best for you to see is, is the way in which we’re kind of experiencing reality right now. We know from behavioral science that humans, although brilliant, tend to be cognitively miserly.

If it’s too hard to think about, we’ll go with something easier. And that becomes the area in which this influence is really changing the way in which we view our worlds. For me, when I think about social media, I think about the fact that I have family members all over the Earth. And the news that they’re reading about what’s going on in one country, uh, colors their view, not only of that country, but of what my experience is, the kinds of conversations we have, what they think is important to me. And so there’s a path dependency that happens due to this intervention that wouldn’t have happened before. At the same time, I’ve been able to, to plan a family reunion via social media, right? So we have this, we have this, uh, dynamic that is tremendously powerful in positive ways, but also is changing the way in which we interact with each other and not always in good ways.

I think, systematically, we need to think about the processes. So these algorithms ostensibly, are meant to improve our decision making, right? To help us be our better selves. And so I think as organizations create these processes that are algorithmically mediated, the first systematic question is: a) Should we do it? Should we? Should we have an algorithm or machine learning making the decisions with us at this point?

The second question, I think we should ask, is to what degree is this being designed with humans in mind? 

And that’s really important because humans are fallible, have biases, and may rely on algorithms to rationalize decisions that otherwise they’d have to explain themselves.

In other words, if we have biases in algorithms and machine learning, and that supports our own biases, we may feel more robust in the flawed conclusions that we make. And so, I think a really important question is how does this interact with the tendencies of human beings to be cognitively miserly, but also to want to be shown that the way they think is right, to not want to be challenged?