Blog Post 6- implications within algorithms
Algorithms have become an integral part of our lives, and they are increasingly impacting our daily routines in ways that we may not even be aware of. These powerful mathematical formulas are used to make decisions and predictions about everything from our shopping behaviors to our job prospects to even what we view online within a day to day basis. Check out this crash course video about algorithmic bias to further understand the topics I am going to discuss throughout this blog.
However, as algorithms become more pervasive, there are growing concerns about the potential for biases and discrimination. One of the most pressing issues with algorithms is the potential for racial bias. Many algorithms are trained on historical data, which means that they may encode and reproduce systemic biases. For example, facial recognition algorithms are less accurate for people with darker skin tones. This can have serious implications for things like law enforcement, where inaccurate facial recognition could lead to wrongful arrests or even violence. Check out this article about "The racist algorithm" to understand what I mean by algorithms being racist.
https://heinonline.org/HOL/LandingPage?handle=hein.journals/mlr115&div=42&id=&page=
Another area where algorithms can be problematic is in advertising, particularly on social media platforms like Facebook. Ad targeting algorithms may inadvertently discriminate against certain groups, such as women or people of color, by showing them different ads or charging them different prices. This can perpetuate harmful stereotypes and reinforce existing inequalities. Within a society that is trying to change and correct systematic biases including protests like Black Lives Matter, this can be very consequential. Look at the video linked below for more information about the "timeline of gender bias in AI"
As well, these biases could teach the younger generation about biases and racism making them racist or believe in inequalities because online platforms have been showing them this from a young age. Another issue that arises through algorithmic biases and implications is shadowbanning. For anyone on Tik Tok, you would know about creators not being shown to many people because algorithms try to hide them. Therefore shadowbanning is limiting the visibility of certain users or content on social media platforms without notifying anybody. This in a way is a form of censorship for marginalized voices. Which leads back to the idea of biases within the system of the internet, creators, protests, and ideas are censored to many people as the algorithms are in a way racist. Check out this Ted Talk about keeping algorithmic bias out of algorithms.
We must also touch upon the topic of AI Ethics, for those of you that do not know AI is artificial intelligence, so robots in a way. AI the past year or so has become very powerful and sophisticated and has the potential to transform industries and improve our lives in countless ways. Algorithms and AI are powerful tools that people do not recognize often, they live behind the closed doors of the internet and create biases and discriminations that are embedded into childrens and adults minds. The article linked below further discusses the impact of AI and algorithmic discrimination.
It is hard to say if this issue could ever be fixed, as we would have to fix and racist actions, discriminations, and more through our society which could never be changed due to our past. Yet, through constant engineering I believe that we could change AI to focus more on our present societal actions than our past and slowly reverse the implications within our algorithms that live behind our internet usage. Tune in next week to learn more about the internet and all the crazy things that come along with it.
Comments
Post a Comment