NLP and ML: Streamlining BofA’s Document Tagging Process; The Ethics Behind Artificial Intelligence

This past summer I interned at Bank of America (BofA) as a Software Engineer Intern, where I worked on an AI/ML team. My team focused on creating machine learning models and applications that automated manual processes. I was tasked with streamlining the tagging process of company profile documents on BofA’s Global Corporate Investment Banking (GCIB) Launch platform. This was a very tedious and inefficient process which took around 10 minutes for someone to read the document and tag it one at a time. It would take an exceptionally long time to tag over tens of thousands of documents if done manually. Hence, I was assigned to create an automated pipeline in Python that would tag ten key attributes from a company profile document. For this project, I used a machine learning model and natural language processing (NLP). This project is what I wrote about in further detail in my technical report. It is also what influenced my topic chosen for my STS Research Paper. Although the project itself was simple and straightforward, it really opened my eyes to how AI and ML are being used in the workforce to improve efficiency and eliminate mundane tasks. Hence, I decided to do my STS Research Paper on the ethical dilemma that comes with the advancement of AI.

In the research paper, I outlined the pros and cons of the advancement of AI and did an analysis of the different social groups that play a role in its advancement using the Social Construction of Technology (SCOT) framework. The goal of this research paper was to decide whether the benefits outweigh the potential drawbacks when it comes to AI. I concluded that to reach a state of stabilization where AI is more beneficial to society and meets all stakeholders demands, policymakers are the group that hold the most power to make an impact. To reduce the ethical implications that come with the development and use of AI, policymakers can set forth certain guidelines’ companies must follow if they want to develop a form of AI to prevent potential ethical issues affecting the users.

Overall, both the Technical Report and the STS Research Paper were great learning experiences. Being able to implement and use machine learning on my own for the first time was extremely exciting. But also, viewing this topic of controversy from an ethical perspective was eye-opening as developers and companies often overlook it. I hope that one day policymakers can take initiative to enact ethical laws that companies must abide by when it comes to AI so that it is no longer considered a controversial topic because it is such a revolutionary form of technology that has already changed our world in many ways.

School of Engineering and Applied Science
Bachelor of Science in Computer Science
Technical Advisor: Briana Morrison
STS Advisor: Travis Elliott