With this cybersecurity project, we took a closer look at how the detection of SQL injection (SQLi) attacks could be improved. These attacks specifically involve the injection of malicious SQL code into an application's database. From our results, we found that utilizing contextualized word embedding methods achieved over 99% in terms of consistency across various classification algorithms and reduced model training time by 31 times, thus further displaying the significance of this method for SQLi attack detection. This research was also submitted to and accepted by The Third Intelligent Cybersecurity Conference (IEEE, 2023).