O’Connell et al. states that “the tool is directing the user” where built-in artificial intelligence (AI) is an embedded technology where “users don’t have to think about how to use it [the technology tool]”. She provides Google as an example of a platform which uses this AI feature, which limits search options based on a “user profile” (2021, module 2.2).
The website supporting The Social Dilemma (2020) documentary states:
Unlike a simple fixed algorithm, AI uses a system of algorithms and can create or modify algorithms without human intervention through the process of continually optimizing for better and better results – often referred to as machine learning (Exposure Labs, 2020).
This description helps to support a deeper understanding of the concept for me, which highlights the notion that the technology is always ‘learning’ for the purpose of ‘enhancing’ the user experience. The search engine builds upon previous search enquiries and as a result, comes to know more about me, offering tailored predictive options and results. In his article, Rowe (2018) provides an interesting example of a search string where this ‘predictiveness’ based on ‘learning’ is noticeable, using key words that have multiple meanings. Rowe is an American national, using American football teams, so I decided to recreate my own similar search.
Initially I typed in “Stadium Australia”, getting pretty the specific and accurate results that I was intending about the sports and entertainment stadium located in Homebush, NSW.
My next search enquiry was “bulldogs”, which provided options about the Bulldogs football team, where I could have alternatively been referring to the animal.
My next search term entered was “dragons” and I gleaned similar football-related results.
However, then in entering “story” next, predictive options provided were tied to my previous search, offering the first suggestion of “story about dragons”.
So, this makes me consider whether this embedded AI technology a good thing for users of online technology. Are the search results ‘helpful’ in the instances where we are after specific information? By virtue of the results being ‘narrowed’ by our own search histories, we are not necessarily offered a full breadth and variety of results.
Adding to this, The Social Dilemma (2020) documentary highlights that not only basic demographic information is collected whilst online, but every search, click, uploaded photo, status update, comment etc. is tracked, which can therefore be used as a tool of manipulation by governments, big business and other interested parties who have access to this user data, with The Social Dilemma website stating that, “The bigger problem is how the data feeding the underlying algorithms is being used to model and predict all of human behavior – giving the highest bidder the ability to influence us at scale like we’ve never seen before from determining elections to sparking revolutions” (Exposure Labs, 2020).
Finally, another issue related to this tailored and personalised AI technology is the ‘addictiveness’ of social media (especially where our preferences are guided by our interest) which leads to further exploration and time spent on electronic devices. This may seem insignificant when it comes to wanting to view something harmless like “cute cat videos” on YouTube, but a Harvard University study has revealed that social media can be both physically and psychologically addictive, having a negative impact on mental health (Hillard, 2021). I would liken it to ‘getting sucked into a vortex’, which can send people spinning out of control!
References
Exposure Labs. (2020). The social dilemma. https://www.thesocialdilemma.com/faqs/
Hillard, J. (2021). What is social media addiction? Addiction Centre. https://www.addictioncenter.com/drugs/social-media-addiction/
O’Connell, J., Derouet, L. & Korodaj, L. (2021). 2. The information environment [Learning module]. ETL401: Introduction to teacher librarianship. Interact 2. https://interact2.csu.edu.au/webapps/blackboard/content/listContentEditable.jsp?content_id=_4449688_1&course_id=_57504_1
Orlowski, J. (Director). (2020). The social dilemma [Film]. Exposure Labs.
Rowe, K. (2018). How search engines use machine learning: 9 things we know for sure. Search Engine Journal. https://www.searchenginejournal.com/how-search-engines-use-machine-learning/224451/