Establishing Boundaries with AI Writing Assistants is Possible. Bans Are Not.

Educators must not forget that they control what AI tools they introduce to their students and the tone they set for the next generation's opinions about the use of this technology.

Lexica: Oil painting of a young man puppeteering a robot, in front of a typewriter.

The panic around AI represents a major challenge to education, and we are seeing policymakers react to that panic with dizzying speed. New York City schools recently announced a ban on ChatGPT, and many other school districts are following their lead. So have many institutes of Higher Learning in Australia. 

Bans are a bad idea. They won’t work because they are easily thwarted and set the wrong tone with students about AI technology. Educators should be the ones to set the tone about how we want our students to use or not use AI in our classroom, not a software company or a district office. These recent bans are all the more concerning considering reports that Microsoft is moving forward with plans to explore integrating OpenAI’s tools in their Office Suite, so it’s crucial to introduce this tech to our students now—warts and all.

How to Establish Boundaries

Depending on how you teach and what you teach, your reaction to integrating AI is going to relate directly to the content and learning goals of your class. If you teach writing as a method of demonstrating knowledge, like testing or research papers, then thinking needs to be original. In this scenario, you want a strong statement against generative AI (which will be difficult to enforce). However, if you teach writing as a part of learning, you can explore incorporating AI as a writing assistant to help students brainstorm, summarize, explore different points of view, etc.

Here’s the framework I developed with the help of Dr. Chad Russell. It’s intentionally vague in certain areas, like “substantial contribution.” That’s something I plan to discuss with my students throughout the semester.

AI-Assisted Writing vs. AI-Generated Writing

With the rise of AI writing assistants, students must take special care to ensure that they use this new technology ethically and honestly. In our class, we will distinguish between 'AI-assisted writing' versus 'AI-generated writing'. AI-assisted writing is only permitted in this course provided a student uses an AI writing assistant as a collaborative tool to help the student with the development and advancement of their own writing process. Collaborating with an AI writing assistant can include brainstorming, outlining, and drafting, so long as there is substantial writing, research, and composing by the student that is not generated solely by the AI.  'AI-generated writing' means there has been little or no involvement from the student as an author, with the majority of the writing being generated by an AI. The goal of using AI-assisted writing in this class is to help students develop their writing process and critical thinking, not to replace or substitute for either. Therefore, using an AI to generate writing or compositions without substantial original contribution from a student is neither acceptable nor allowed. 

Be Aware:

• You will not be able to save anything the AI generates. If you want to use an idea or suggestion produced by the AI, you will need to copy and paste it into a Word.doc. 

• The AI output may contain material that is offensive, biased, or otherwise goes against the University Creed in practice, or material that is false or misleading or potentially harmful, or other problematic material the use of which may fall outside the protections of Academic Freedom and/or Free Speech--review all AI output carefully before using anything suggested by the AI within your academic work.

Others are also establishing their own framework with their students. Lance Cummings shared a draft of his AI “honor contract” that he plans to use with his writing students.

iSophist 🚢 🤖🏛️ | AI & Digital Storytelling @LanceElyot

Here is the first draft of an AI "honor contract" for my AI & Digital Storytelling class. Students will give input and make changes as we learn more about AI writing together. I'd love to hear anyone's thoughts, though. ⬇️

Explore Student Impressions of AI

At some point, we’re going to have to move beyond panicking about this technology and what we think AI will do to our students and begin to pragmatically approach AI’s impact on education. This means showing your students the technology and talking to them about it.

Integrating AI into your teaching isn’t going to destroy your student’s critical thinking skills, nor will students use it to generate all the writing within the course. Some students will like the tools and others will not. Most students will respect the boundaries you establish, while some will not. These are normal reactions, and if we pause and stop letting panic drive us, we can use this moment as an opportunity to introduce students to the benefits and limitations of this technology, before they learn it through social media, their friends, or some tech company.

Paul Fyfe has used GPT-2 and GPT-3 powered tools with his writing students for several semesters and collected student reflections.

Paul Fyfe / @pfyfe@hcommons.social @pfyfe

Seems like a problem that much journalism and punditry on AI and student writing ... doesn't actually ask the students. They're experimenting and taking away different lessons: +

Paul Fyfe / @pfyfe@hcommons.social @pfyfe

“Writing an AI generated paper was far more difficult than I would have expected.” “I would compare collaborating with [GPT-3] to being teamed up with the class ‘slacker.’” “I feel as though it would be much more effective for me to just write alone.” +

Our Reactions Matter

I know it doesn’t seem so, but we actually have an incredible amount of power in this given moment. The tone we set with our students will likely cement their view of this technology, and if their first encounter with it is someone in authority telling them that an AI writing assistant is only a mechanism to cheat, then that will be how an entire generation of students views the use of this technology as it relates to writing. Instead, let’s focus on nuance and open discourse.

Wikipedia was a recent moral panic in education, and our reactionary approach to this tool still ripples throughout educator’s views on it. As an undergrad, my opinion of Wikipedia was largely shaped by what my professors told me about it—don’t use it, it’s unreliable, anyone can edit it, not in my class. That initial impression shaped my thinking as a student and carried over as a teacher. It’s taken me years to unlearn and unpack the reasoning that shaped the initial panic over something so benign in comparison to our current moral panic over AI writing assistants.

Let’s not make the same mistakes. Educators should use the moment to introduce AI to our students carefully. Ask students to reflect on how using these tools impacted, changed, or challenged their writing process. Talk to them about boundaries that you established and ask them about what limitations they’d like to establish in their own use of these tools. There’s no perfect scenario here and we will all make mistakes, but to do so openly with our students is the path forward.

Resources

• Tru Digital Detox: Ground Rules for the Robot Wars: Defining Our Terms

• SFCC Library: Faculty Help: ChatGPT's Impact on Higher Education: Resources: FAQ

• Faculty Focus: What are we doing About AI Essays?

For Your Listening Pleasure