Algorithmic Bias and Filter Bubbles
While researching on algorithmic bias I found a very insightful TED Talk by Eli Pariser about Filter Bubbles. Pariser says that Filter Bubbles are “your own personal, unique universe of information that you live in online. And what's in your filter bubble depends on who you are, and it depends on what you do. But the thing is that you don't decide what gets in. And more importantly, you don't actually see what gets edited out.” (Pariser, 2011). He describes in his talk how algorithms define what gets filtered in or filtered out of our information feeds.
A good example of this filtering is what we experience in social media. Algorithms decide what to show and what not to show. This decision making seems to be neutral, but as I have discussed in previous posts, no algorithm is neutral given that they are only mathematical representations of a set of particular human views of what should be prioritized and what not.
I found his talk very compelling and revealing. Making a brilliant analogy during the talk, Pariser unravels the sometimes-complex topic of algorithmic decision-making by saying: “What we're seeing is more of a passing of the torch from human gatekeepers to algorithmic ones. And the thing is that the algorithms don't yet have the kind of embedded ethics that the editors did.” (Pariser,2011).
Adding to the conversation, I believe that Filter Bubbles are the result of algorithmic bias. Designers who only work on maximum engagement as the ultimate goal of their algorithms but set aside important concepts, like balance and ethics, may only be injecting dangerous human bias towards immediate reward into their code. We all know that balance and ethics are important. We know that too much of the same thing rarely leads to anything good. Would perhaps a more human, ethical version of these algorithms be one that avoids filter bubbles and provides a more balanced result? Understanding what biases are present in our code may be a good starting point.
References:
Pariser, Eli. (2011, March). Beware online "filter bubbles" [Video]. TED Conferences. https://www.ted.com/talks/eli_pariser_beware_online_filter_bubbles/transcript?language=en
Beware online "filter bubbles" (source: TED Talks)
A good example of this filtering is what we experience in social media. Algorithms decide what to show and what not to show. This decision making seems to be neutral, but as I have discussed in previous posts, no algorithm is neutral given that they are only mathematical representations of a set of particular human views of what should be prioritized and what not.
I found his talk very compelling and revealing. Making a brilliant analogy during the talk, Pariser unravels the sometimes-complex topic of algorithmic decision-making by saying: “What we're seeing is more of a passing of the torch from human gatekeepers to algorithmic ones. And the thing is that the algorithms don't yet have the kind of embedded ethics that the editors did.” (Pariser,2011).
Adding to the conversation, I believe that Filter Bubbles are the result of algorithmic bias. Designers who only work on maximum engagement as the ultimate goal of their algorithms but set aside important concepts, like balance and ethics, may only be injecting dangerous human bias towards immediate reward into their code. We all know that balance and ethics are important. We know that too much of the same thing rarely leads to anything good. Would perhaps a more human, ethical version of these algorithms be one that avoids filter bubbles and provides a more balanced result? Understanding what biases are present in our code may be a good starting point.
References:
Pariser, Eli. (2011, March). Beware online "filter bubbles" [Video]. TED Conferences. https://www.ted.com/talks/eli_pariser_beware_online_filter_bubbles/transcript?language=en
Comments
Post a Comment