1 of 46

Introduction to PapyrusAI

Daniel Ritchie and Dana Saito-Stehberger

2 of 46

Today’s Agenda

  • Introductions
  • Overview of the PapyrusAI Pilot Project
  • What is the difference between ChatGPT and PapyrusAI?
  • Demonstration: How PapyrusAI works
  • Tinkering with PapyrusAI
  • Resources
  • Questions

Digital Learning Lab

3 of 46

Introducing Ourselves

  • What do you know about Generative AI?
  • In what ways have you used Generative AI?
  • How can you imagine using Generative AI in your classroom?

Digital Learning Lab

4 of 46

Mark Warschauer

Tamara Tate

Daniel Ritchie

Dana Saito-Stehberger

UCI Team

K-12 Lead

Curriculum /

Teacher Support

Kristi Werry

Software Engineer

This material is based upon work supported by the National Science Foundation under Grant No. 23152984.

© 2024 The Regents of the University of California

Gerardo Jr. Lopez

Research Assistant

Digital Learning Lab

5 of 46

Aims of the PapyrusAI Project

  • To learn how Generative AI can be used to help students
  • To learn how tools can be designed to support teachers and students in the classroom

Digital Learning Lab

6 of 46

Using Papyrus AI

  • You will develop a better understanding of what Generative AI is and how to interact with it.
  • You will be among the first teachers to integrate AI into your teaching and instructional strategies.
  • Your students will be given the opportunity to interact purposefully with Generative AI.
  • Your feedback will help us to continue to develop this tool to benefit teachers and students in the classroom.

Digital Learning Lab

7 of 46

What is a large language model (LLM)?

ChatGPT and other large language models (LLMs) are a type of artificial intelligence that are trained to generate text similar to human-generated text through text prediction. They cannot think or reason. They are called “large” because they are typically trained on a corpus of text data ranging from millions to billions of words.

Digital Learning Lab

8 of 46

Data Input

Some Tasks

Text summarization

Text generation

Sentiment analysis

Content creation

Virtual assistants

Speech Recognition

Image annotation

Text-to-speech

Spell correction

Translation

Training

Adaptation

Digital Learning Lab

9 of 46

What is the difference between ChatGPT and PapyrusAI?

LMS + ChatGPT 4o

  • Gives teachers control of how students interact with AI
  • Provides scaffolding and support as students develop their writing skills and process

Digital Learning Lab

10 of 46

How has PapyrusAI been used in middle school classrooms?

  • As a thinking partner as students wrote narrative and argumentative essays
    • Brainstorming ideas
    • Guidance in the writing process
    • Feedback on writing
    • Encouragement

Digital Learning Lab

11 of 46

The Prompt

Digital Learning Lab

12 of 46

. . .

Digital Learning Lab

13 of 46

Digital Learning Lab

14 of 46

Digital Learning Lab

15 of 46

Digital Learning Lab

16 of 46

How has PapyrusAI been used in middle school classrooms?

  • Students learned more about the historical background of the Holocaust as they read Number the Stars. The teacher said,

It was really cool because it was just last minute. We usually do a reading and then they answer questions and learn a little bit about the Holocaust. I was like, why not have them just ask stuff about it? . . . Especially, in honors class, there's very different levels of what kids might know about World War II or the Holocaust. You always have your very history-wiz kids that know more than I do. Then, there will be some kids that have vaguely heard about it but don't know much. I think it was really good to do it that way because it was adaptive to how much they knew. It would ask them back like, "Have you heard of this?" If they said no, it would explain it, and if they said yes, it would go more into detail, more in depth. I really do think they enjoyed it.”

Example Student Instruction Sheet

Digital Learning Lab

17 of 46

What is generative AI good at?

  • Brainstorming/thinking partner
  • Writing text
  • Feedback on text
  • Style of text

Start with the easy ones!

Digital Learning Lab

18 of 46

What is generative AI good at?

  • Brainstorming/thinking partner
    • Topics
    • Counter arguments
    • Evidence
    • Perspectives
    • Rubber duck debugging

Think first!

Often part of planning, pre writing

Tamara

Digital Learning Lab

19 of 46

What is generative AI good at?

  • Writing text
    • if truth doesn’t matter or
    • you know enough to gauge accuracy AND
    • writing the text is not part of the learning objective

Hallucinations

Learning objectives

Tamara

Digital Learning Lab

20 of 46

Writing text

  • Where does “truth” not matter?
  • Where are students experts or have access to the source material so they can corroborate the accuracy?
    • Summarizing a source
    • Creating a quiz on content (textbook, instructor slides) the student needs to know, for help reviewing
    • Summarizing the student’s writing

Tamara

Digital Learning Lab

21 of 46

What is generative AI good at?

  • Feedback on text
    • On its own or
    • In conjunction with peer review

Hallucinations

Learning objectives

Opportunity to motivate student revision

Tamara

Digital Learning Lab

22 of 46

Large Language Models:

Therefore,

LLMs should be used when:

Users should remember:

  • Are built on prediction & language.
  • Don’t think.
  • Have no core beliefs.
  • Fabricate things rather than say “I don’t know”.
  • Were trained on a specific set of texts & by humans, so very real biases.
  • Truth doesn’t matter.
  • You know enough to gauge accuracy.
  • That AI doesn’t think or understand. It does not have opinions and cannot make value judgements. It’s just predicting the next bit of text given what you told it and what it has been trained on.
  • NOT to put sensitive personal information or confidential information into any AI tool.
  • AI hallucinates and has bias: Users are responsible for the accuracy and appropriateness of anything you use or incorporate into your own text.

Digital Learning Lab

Digital Learning Lab

23 of 46

Getting started with PapyrusAI

Digital Learning Lab

24 of 46

Creating your account

Resource: PapyrusAI Guide

Any issues with login or access?

Digital Learning Lab

25 of 46

Joining a course

  • How will you manage the process of students setting up account & joining course?

Code to join the class:

sandboxfun

Digital Learning Lab

26 of 46

PapyrusAI.org

Digital Learning Lab

27 of 46

Digital Learning Lab

28 of 46

Digital Learning Lab

29 of 46

  1. Try ‘Planning’ for 5 minutes.
  2. Title ‘Planning_YourFirstName’

Digital Learning Lab

30 of 46

  1. Try ‘Revising’ for 5 minutes.
  2. Title ‘Revising_YourFirstName’

Digital Learning Lab

31 of 46

  • Try ‘Writing a Story’ for 5 minutes.
  • Title ‘Story_YourFirstName’

Digital Learning Lab

32 of 46

Create a course

Digital Learning Lab

33 of 46

Downloading a Conversation

Digital Learning Lab

34 of 46

“Deleting” conversations

Delete

Note: This only hides the conversation from your view. The instructor of the course can still access deleted conversation.

Digital Learning Lab

35 of 46

Edit a course & create module(s)

You can duplicate courses too!

Digital Learning Lab

36 of 46

Selecting prompts

  • Pinned prompts were created and tested by research team
  • You can create your own folder
    • And your own prompts
    • Or edited versions of our prompts
  • ONLY the prompts you select will be available to your students

Digital Learning Lab

37 of 46

Create your own folder

Folder

Digital Learning Lab

38 of 46

Create your own prompt

From the folder you want it in

Digital Learning Lab

39 of 46

Additional features

Digital Learning Lab

40 of 46

PapyrusAI prompts

We have grouped the embedded prompts into the following categories:

  1. Planning
  2. Drafting
  3. Revising
  4. Presenting
  5. Instruction
  6. Miscellany
  7. Retired Prompts

Tamara

Digital Learning Lab

41 of 46

Tags:

Recommended = Researcher created and tested

Expand for view of full text of prompt

Digital Learning Lab

42 of 46

Important points to keep telling students

  • AI can’t think
  • AI makes stuff up, you have to check it
  • The best way to keep stuff private is to not put it into the AI
  • You are the author, you decide what to implement
  • You are the author, you have to make sure your work is accurate
  • What bias or limitations might be impacting the AI output?

Digital Learning Lab

43 of 46

“Getting Started” Lesson Plan

Digital Learning Lab

44 of 46

Resources to Support You

Digital Learning Lab

45 of 46

Questions?

46 of 46

Thank you!

Daniel Ritchie and Dana Saito-Stehberger

This material is based upon work supported by the National Science Foundation under Grant No. 23152984.

© 2024 The Regents of the University of California