In November 2025, the Martin V. Smith School of Business & Economics launched its first issue of Channeling AI: A Newsletter of AI and Business Analytics for the CI Community. To spotlight this achievement and provide some context, we had a Q&A with Minder Chen, Professor of Management Information Systems, and Nicholas Bridgman, Graduate Assistant AI Researcher and Masters Candidate in MS Business Analytics. We invite you to read the Q&A and then to read our re-print of the newsletter, posted with permission of MVS.
How Did the Channeling AI Newsletter Come to Be — Impetus and Origin?
Response provided by Dr. Chen
The Channeling AI Newsletter grew directly out of the AI and Business Analytics Lab (AiBAL), led by Professors Minder Chen and Hua Dai at the Martin V. Smith School of Business & Economics, funded by the M. Scott Funding for Existing Academic Programs initiative at CI. From the outset, the lab planned to publish one or two newsletters per semester during the funding period to capture case studies; student and faculty projects; regional business uses of AI; and innovative teaching practices using generative AI.
What Do You Hope to Achieve with the Newsletter?
Response provided by Dr. Chen
Conceived as the lab’s communication arm, the newsletter showcases applied AI and business analytics learning, teaching, and research, while amplifying the impact of AiBAL’s seminars and workshops. It aims to:
- Build a cross-campus and regional community of practice—faculty, students, and industry—by extending the reach of AiBAL workshops, seminars, and guest talks through accessible summaries.
- Strengthen industry partnerships and workforce readiness by highlighting local collaborations and applied analytics work aligned with regional needs.
- Raise CI’s visibility as a leader in AI-integrated business education by consistently sharing evidence of impact and best practices from across campus and the region.
Together, the AiBAL grant and the M. Scott Funding initiative drive the newsletter’s purpose: to convey the latest developments in AI across CI through a publication that positions CI graduates as leaders in AI-integrated business.
Throughout the newsletter, you use AI transparency statements that normalize the use of AI in the composing process. Can you tell us more about your design & composing decisions?
Response provided by Nicholas Bridgman
I designed a “Rules” prompt containing dozens of the heuristics I follow when writing prose, which I uploaded to Files in a ChatGPT Project. This directed it to follow my prompt’s guidelines every time it responded to a query. I created the prompt based on Dr. Chen’s suggestion of first writing down all the writing techniques I learned as a Rhetoric major at U.C. Berkeley, such as avoiding passive verbs, keeping clauses and sentences concise, and using argumentative structure. Then I iterated with the AI, entering queries, critiquing where I felt its responses needed improvement, and receiving its feedback on how I could alter the prompt to reflect my critique. Over the hours I spent working with it, I developed a prompt with multiple sections, each with dozens of rules, including:
- A Permanent Enforcement Tier, with non-negotiable pre-generation constraints, including thesis integrity, sentence construction and rhythm, and paragraph architecture.
- An Advisory Tier, with contextual and stylistic guidance to shape voice but not override the Enforcement Tier, with rules and examples for sentence structure, word choice variation, and weak and strong diction.
- A Final Verification section, directing it to perform several checks and corrections before finalizing any output, such as using “flag” words appropriately to structure paragraphs, maintaining natural subject-verb order, and ensuring precise and natural rhythm and meaning.
After hours of iteration, I perfected this prompt so that ChatGPT consistently produced text incorporating all my guidelines. By following every rule I could think of to codify my rhetorical style, it generated prose almost indistinguishable from what I would write. While I generally like to write myself, due to being very detail-oriented about what I want to say and how to say it, this AI generation could prove useful if I did not have time to draft a full document, but wanted to have something in my voice. It could also help in cases where I have taken notes on a lecture or subject, and I would like them converted into a coherent essay that captures the key ideas in my writing style (as in the panel discussions in the newsletter). I feel this represents a reasonable use of AI, because although I did the work of taking the notes, turning them into an article would previously have been too tedious and time-consuming to do very often. But since I trained AI to do this so efficiently, it makes sense to take advantage of the technology available, as long as I am transparent about what I use it to do.
Newsletter
This newsletter has been reprinted in the Teaching & Learning Innovations blog with permission from the Martin V. Smith School of Business & Economics.
