Humans and Automation

Introduction: I am just completing my last week of classes for my first semester at Grinnell College. I’ve had a great time and learned a lot about a wide variety of topics. Here is a paper that I had to write for my tutorial class which was on automation. I felt this paper was worth sharing as it is a topic that is very important to our society yet few people are talking about. I would love to hear your thoughts on this paper and the topic in general. I also encourage you to read Automating Inequality by Virginia Eubanks. It is an eyeopening book on the creation of the “digital poorhouse.” Come to your own conclusion and formulate your thoughts on your own. I hope you enjoy and have a great holiday season.

In Automating Inequality Virginia Eubanks documents critical evidence on the flaws of many automated systems that are being used to manage states’ social programs. Though she makes many important discoveries and eye-opening realizations, she is unwilling to recognize the positive impacts that an automated system, if operated properly, could have on people and society as a whole. One of the main systems she focuses on, the automated eligibility system in Indiana, was meant to automate the welfare eligibility system, to cut costs and streamline efficiency. Due to poor implementation and a negligent government the system was a complete disaster. However, Eubanks falsely interprets this lack of success as evidence that all automated systems will lead to problems. In my opinion, it was not the idea that was the problem- in fact automated systems for welfare have the potential to transform social programs for the better- it was the implementation of the system that led to people not receiving entitlements, the loss of critical documents, and the complete breakdown of the government to provide assistance to those in need. It takes time for investments in future technologies to pay off, and hiccups or mistakes will inevitably be made along the way, yet automating systems in our government agencies that manage social programs is an investment worth making. It is important to recognize and learn from the problems that an automated system creates; nonetheless, innovative technologies can enhance the capabilities of these systems and provide the greatest number of resources to those in need. Over the long term, automated systems used to manage states’ social programs will lead to more efficiency, less fraud, and greater outcomes for the people who need help the most.   

When automated systems become too impersonal, they can lose the very important human aspect required by social services. It is important to ensure that algorithms are not the only deciding factor when it comes to who gets benefits. However, even Eubanks admits that something needs to change because the systems we have now, across the country, do not meet the needs of vulnerable people in our society. Eubanks states “Cumbersome administrative processes and unreasonable expectations kept people from accessing the benefits they were entitled to and deserved” (Eubanks, pg. 179). A change to the systems that manage social programs is long overdue, yet the solution to this problem is not to eliminate automation; the solution is to combine automation with the humanity of a caseworker so as to better increase the outcome of clients.   

In the case of Indiana and the automated eligibility system, the way in which it was implemented was terrible. However, the intentions of the Governor and the private companies who helped build the system were good, in fact as Eubanks states the goals of the system were to “reduce fraud, curtail spending, and move clients off the welfare rolls.” (Eubanks, pg. 46) Unfortunately, the consequences of the automated system hurt far too many vulnerable people. In the past a caseworker would visit a client for years and get to know them on a personal level. When the new system was enacted, this changed, and caseworkers were told not to get to know their clients so as to better improve efficiency. On top of this three major changes were made, to the system, that had significant consequences. The eligibility system was turned into a telephone-based system, online applications were required, and files were digitized. This led to a 2000 percent increase in the error rate of the system as well as the loss of 283,000 documents (Eubanks). Complaints went through the roof, clients could not get resources they desperately needed, and worst of all, litigation began against the state and the companies who created the system. In the end, a promising idea turned into a total disaster.  

The new system was a complete failure. Thus, many people might argue that the old system was better, and automation led to this disaster in Indiana. However, automation was not the problem. Given that, the idea of an automated system was good, and the implementation of that system was poor, if the process of creating a new automated eligibility system in Indiana could be done again here is how it should be implemented. First of all, the implementation of this system needs to be slow and gradual over time. People and even machines and algorithms do not respond well to quick and sudden changes. By implementing the automation slowly, you would allow both caseworkers and clients to get used to the new system.  

The second way to improve the algorithm is to increase the amount of resources given to caseworkers. The biggest two complaints with the system in Indiana were that caseworkers were overburdened and untrained with how to use the system and secondly that it was more difficult for clients to access their benefits. As Eubank says, “Private call center workers were not adequately trained to deal with the severity of challenges faced by callers.” (Eubanks, pg. 50) By giving caseworkers more resources, we allow the algorithm to work with people not against them. An important question to ask is, do state governments have the ability and funding to implement a better system and give caseworkers more resources? Though it is true that at first state governments will have to spend more money to improve the system. In the long run the investment will pay off as resources are given more efficiently to those in need. This in turn will decrease the amount of people in need and lower the amount the state needs to spend on social programs. Moreover, a world in which machines and people work together will increase efficiency and outcomes in all areas, specifically, states social programs.  

Eubanks sheds light on many issues that come about when implementing new technology in an old system. The topics she addresses revealed many problems in our government, society, and the overall way that America treats the poor. The automated eligibility system in Indiana is a perfect example of how an inefficient and incompetent government failed at using automation to improve the lives of its citizens. Automation itself was not the problem. The implementation of the automation was what led the new system to fail. What this book reveals is that, in the United States, there are some serious flaws in how we get resources to those in need. The old way states used to manage their social programs was clearly broken. The automated eligibility system was an attempt to fix a system that had failed so many. Unfortunately, the new system caused much more harm than it did good. To blame automation for the failure of the new eligibility system in Indiana is not reasonable. What the automated eligibility system in Indiana showed was that automation and technology alone cannot fix an entire system. Instead, humans need to use automation as a tool, not the other way around. 

References:  

Eubanks, V. (2019). Automating Inequality: How High-Tech Tools Profile, Police, And Punish The Poor. St Martin’s Press  

Previous
Previous

The Fallacy of American Democracy

Next
Next

Levels of Analysis and Hierarchy in International Relations