This research examines how Large Language Models impact software testing effectiveness and productivity in real-world engineering contexts.

Key Findings:

• LLM-assisted unit testing significantly improves defect detection rates
• Developers using LLMs complete testing tasks with greater efficiency
• The quality of test cases improves when engineers leverage AI assistance
• LLMs help overcome common testing barriers, particularly for complex systems

For engineering teams, this research demonstrates how AI can augment human capabilities in critical quality assurance processes, potentially reducing bugs in production while accelerating development cycles.

Unit Testing Past vs. Present: Examining LLMs' Impact on Defect Detection and Efficiency