Can Artificial Intelligence Pass a Sophomore Level Digital Design Laboratory?
Is this project an undergraduate, graduate, or faculty project?
Undergraduate
Project Type
group
Campus
Daytona Beach
Authors' Class Standing
Casey Elder, Sophomore Alessia Tripaldelli, Sophomore George Pozek, Sophomore Samuel Horine, Junior
Lead Presenter's Name
Casey Elder
Lead Presenter's College
DB College of Engineering
Faculty Mentor Name
Brian Butka
Abstract
Artificial intelligence technology is rapidly growing new capabilities with the introduction of natural language processing and generative pretrained transformer models such as chatGPT3. This research investigates whether this model can complete a sophomore level digital design laboratory. This work took a set of existing laboratories for the class CEC222 at Embry-Riddle Aeronautical University and tasked ChatGPT with completing while not being able to physically wire the circuits it was able describe the wiring, write the programs required to execute the laboratories and answer the laboratory questions. When the labs were independently graded, ChatGPT achieved a grade of 73%. This paper describes the labs, gives examples of the results, and discusses the successes and failures of ChatGPT as applied to this laboratory class. The paper concludes with a discussion of ways that ChatGPT can be integrated into the curriculum.
Did this research project receive funding support (Spark, SURF, Research Abroad, Student Internal Grants, Collaborative, Climbing, or Ignite Grants) from the Office of Undergraduate Research?
No
Can Artificial Intelligence Pass a Sophomore Level Digital Design Laboratory?
Artificial intelligence technology is rapidly growing new capabilities with the introduction of natural language processing and generative pretrained transformer models such as chatGPT3. This research investigates whether this model can complete a sophomore level digital design laboratory. This work took a set of existing laboratories for the class CEC222 at Embry-Riddle Aeronautical University and tasked ChatGPT with completing while not being able to physically wire the circuits it was able describe the wiring, write the programs required to execute the laboratories and answer the laboratory questions. When the labs were independently graded, ChatGPT achieved a grade of 73%. This paper describes the labs, gives examples of the results, and discusses the successes and failures of ChatGPT as applied to this laboratory class. The paper concludes with a discussion of ways that ChatGPT can be integrated into the curriculum.