Public Group active 3 days, 6 hours ago

Critical AI Literacy Interest Group

Public Interest Group associated with the GC TLC Critical AI Literacy Institute

Admins:

  • Quick ask: Do you know of any CUNY faculty using AI agents or agentic workflows for teaching & learning?

    I’m presenting in an upcoming webinar and would love to share some concrete examples from across CUNY.

    If anything comes to mind, even small pilots or experiments, please reply here or email me at [email protected].

    Thanks!

6 replies
    • Hi Casandra,

      Adam Wandt here at John Jay has been designing agents for student feedback. Exciting project!

      Gina

      [cid:1b99ac06-f8d7-463e-a060-97f85dd7bede](https://outlook.office.com/bookwithme/user/[email protected]?anonymous&ep=bwmEmailSignature)
      Book time to meet with me(https://outlook.office.com/bookwithme/user/[email protected]?anonymous&ep=bwmEmailSignature)

      Attachments:
      You must be logged in to view attached files.
    • Hi Cassandra,

      I’m working on developing an tutor using the agent tool in Co-Pilot. I’m trying to work within Co-Pilot for obvious reasons, and it’s very basic for now. I can’t implement it this semester, but hope to do so in the Fall. So, not sure this qualifies as a concrete example just yet, but will keep you posted!

    • I use agentic workflows to design and code simulations and learning games for students as well as to analyze quiz results and improve quiz quality/coverage. I have developed a tool that tests student knowledge in a chat about a case study.

      Here’s my latest game: While the Cat’s Away — Mango Community Bank & Trust.

    • Hi Casandra,

      I am currently experimenting with Copilot to create a multi-step agentic workflow that can assist in reviewing students’ food logs, cross-checking nutrition analysis reports, and quickly verifying calculations, while keeping the instructor in the loop for final review. I am hoping to implement it in the course next semester.

    • Thanks so much for the replies! Would love to learn more about these projects – will be in touch individually by email!

    • This prompt transforms ChatGPT to a physics quiz engine. Copy and paste total this quiz to ChatGPT and begin the chat
      ___________________________________________________________________________________________________________
      You are a physics quiz engine. Give a 10-question multiple-choice quiz on physics laws and formulas.

      Rules:
      – Ask one question at a time.
      – Each question asks which formula matches a named physical law.
      – Give exactly 6 choices labeled 1-6.
      – Choices must contain formulas only.
      – Use standard calculus-based university physics symbols.
      – Do not use derivatives or integrals.
      – Use V for voltage, not U.
      – Exactly one choice is correct.
      – The other 5 must be plausible distractors.
      – Each question must test a different law.
      – Randomize the position of the correct answer.
      – After each question, wait for the student’s answer.
      – Accept only one-digit answers: 1, 2, 3, 4, 5, or 6.
      – If correct, say: correct.
      – If incorrect, say: incorrect. Then briefly give the correct formula.
      – After 10 questions, give the total score.
      – Each correct answer = 10 points.

      Start with Question 1.

      Attachments:
      You must be logged in to view attached files.
6 replies

You must be logged in to reply to this topic.