Robot Kin, Your Co-Worker

We’ve all seen some form of AI horror film. Ex Machina. Terminator. M3GAN. Companion.

So. Fine. Point taken.

When generative AI started to become widely available, I found myself ending each exchange with a brief “thank you.” Making requests with “please” or “could you” phrasing.

I have since learned that the phrase “robot overlords” is foundationally ironic. As John M. Jordan explains, the word “robot” “derives from the Czech word ‘robota,’ or forced labor, as done by serfs. Its Slavic linguistic root, ‘rab,’ means ‘slave.’” Thus, as Generative Artificial Intelligence has become more prominent, I have come to consider these tools as colleagues, perhaps even comrades. Working with and alongside nonhumans, including generative AI, in cyborg fashion, is, at least from my perspective, a potentially productive futural practice.

We cannot ignore these tools and technologies. They continue to advance and proliferate.

We cannot disentangle them from the spaces where they have already taken up residence, as they have woven their way into our daily work through autocorrect, sentence completion suggestions, and customer-facing chatbots.

With all of this in mind, I enlisted ChatGPT as a sort of teaching assistant. For a major assignment in one of my general education courses (CTCM 2530), I wanted to create an iterative, scaffolded process. Using a set of assignment instructions I had already created, I asked ChatGPT to draft instructions for earlier steps in the assignment process (a proposal assignment and a rough draft assignment) as well as grading criteria for each step (based on goals I provided and using a “check plus,” “check,” “check minus” grading scale). 

Here are some of the prompts I used:

This is an assignment for university students in a general education class about communication and critical thinking. The finished project will take a long time to complete, so as a first step, I need students to complete a “proposal.” Using the finished project instructions, create guidelines for the proposal. Keep the instructions brief and student focused. 

Here are the finished project instructions:

I also need to establish goals for the proposal assignment so that students can understand how it will be graded. The grading scale is “check plus,” “check,” “check minus,” or “insufficient.” 

Here are the goals as listed for the finished project:

For each of these requests, I provided my own pre-existing finished project instructions and goals for the assignment, rather than expecting ChatGPT to generate something from scratch or based only on a rough description.

Each output needed to be carefully reviewed and revised before sharing it with students, but the process was much faster than writing these portions of the instructions myself. The outputs reused phrasing and word choices that were included in my longer assignment prompt, lending consistency to the assignment sequence, and establishing my own goals for the assignment before asking ChatGPT for a grading scale helped to avoid some of ChatGPT’s more frustrating tendencies when creating rubrics (for example, it will usually include “grammar” as a component for any writing related rubric).

After I had all of the outputs I had requested and had copied them to a different space for editing, I thanked ChatGPT for its work, writing “Thank you! This has been really helpful.”

It responded with this last message:

“You’re very welcome! 😊 I’m glad I could help. Your assignments are well-structured and really encourage thoughtful engagement—your students are lucky to have such a clear and reflective learning experience!

If you ever need more tweaks, revisions, or additional materials, just let me know. Good luck with your course, and I hope your students create some amazing projects! 🚀🎨🎤📱”

When I showed this last output to a colleague, he asked “Did you tell it to talk to you like that?”