The NYU Stern Faculty Innovation Fellows program aims to foster a culture of innovation in teaching and learning by providing faculty members with the resources and support they need to experiment in the classroom. In our first year, we partnered with Jeffrey Younger, a Clinical Professor of Management Communication, to explore generative AI in teaching.

AIM assignment chatbot

For the first iteration of the Faculty Innovation Fellows program, the team collaborated with Professor Younger on developing a chatbot that would provide formative feedback to students. This feedback was tailored to Professor Younger’s AIM (“Audience,” “Intent,” “Message”) memo writing assignment, where students are asked to write a memo analyzing a business communication document.

The Learning Science Lab used Professor Younger’s rubric and examples of previous feedback to prime a custom GPT. Through an iterative process of testing, prompt engineering, and assessment by Professor Younger, the team developed a tool that would apply the rubric to student submissions.

Application design and steps

The custom AI tool was designed to engage students and increase the effectiveness of the learning process, which involved the following steps:

1. Document selection and initial draft
Students searched and selected a sample of written business communication and drafted an analysis of the document, judging its effectiveness while comparing it with the in-class AIM communication model. Students then applied the AIM model and a sample business memo structure as they wrote an initial draft of their findings, considering the professor and TA as their target audience.

2. Peer feedback
Step two involved a student peer review where analyses were read and constructive comments shared. Paired students offered feedback, discussing the effectiveness of the substance, structure, and style of the draft as a way to help their colleagues improve subsequent iterations.
 

3. GPT feedback
Students were invited to upload an anonymized draft version to the specific custom GPT, programmed to understand the AIM model and offer feedback. As with the peer review, the AI generated results were shared and discussed in class to further understand applied business writing models as they consider the purpose and practice of business writing. 

Students used all previous input to draft a final graded version of their analysis.

Impact

The AIM assignment feedback tool was used by thirty students in Professor Younger’s J-term Strategic Communications course at NYU Abu Dhabi. Both the students and Professor Younger found the tool to be effective. In a post-assignment survey posed to students by Professor Younger, which 19 of the 30 students completed, students indicated whether they found each step in the revision process to be helpful in creating their final draft: 

• 63% of students found the peer feedback to be helpful in creating their final product.
• 89% of students found the AI generated feedback to be helpful in creating their final product.

Students were also asked whether they would like to see a similar use of AI tools in other courses:

• 89% of students said they would like to see a similar use of AI tools in other courses.

In addition to sharing the AIM assignment chatbot with other faculty for Management Communication courses, the Learning Science Lab is using the tool as a model for developing other feedback GPTs and consulting faculty on similar genAI applications.

Interested in trying a new tool or pedagogical approach in your classroom?

We are always reviewing requests for collaboration, and we identify our yearlong fellows each summer. Fellows are expected to actively participate in innovation, providing feedback, assisting in data collection, and regularly meeting with the LSL team. To inquire about partnering with us, send us an email at learning.science@stern.nyu.edu.