This week’s tip: AI use needs to be carefully considered.
This week’s tip is a bit different and a lot longer. I want to share an issue I have enjoyed struggling with for the last two years.
I love tech. We created TimeTrack, a tech tool, in 2005, so principals, and other school leaders could use to use it to improve their impact ~ a central part of the SAM Process.
Two years ago I was excited with the emergence of artificial intelligence, AI. I played with it a bit, but my excitement was tempered by the tool’s hallucinations, totally made up statements it presented as fact. I banned the use of AI by Time Change Coaches, and our TimeTrack developers, until we could develop guardrails for its use. At the same time, we made sure TimeTrack would be able to incorporate AI at some point.
Since then, I have been impressed by SAM principals who use AI effectively. Jill Jacobs, an amazing SAM principal in Missouri, uses AI in a very effective, and ethical, manner:
“I like to think of ChatGPT as adding the "SAM" factor to my written messages. With that I mean it adds the personal relationship piece that I sometimes find hard when I am balancing the various demands on my time. Additionally, using AI to create some of my messages helps me get management tasks done promptly. When I am not sure how to respond to an email I procrastinate, now I can "tell" ChatGPT what I want to say and it composes a response in a tone that promotes my true intentions, care, and connection.”
Pretty cool, right?
Jill’s use, and what I’ve learned other SAM school leaders do, demonstrate effective and ethical use of AI.
Sunday morning, I always ready a hard copy of the New York Times. It is still sitting next to my living room chair this evening as I also received, via email, The Main Idea, a monthly review of a book of interest to educators.
Once I started reading the summary and support materials, I could not go back to the Times. The book? The Artificial Intelligence Playbook: Time-Saving Tools for Teachers that Make Learning More Engaging by Meghan Hargrave, Douglas Fisher, and Nancy Frey.
So, I played with the question that keeps me from using AI: “Ethically, should a leader disclose when they use AI in an email or other document?”
I warned you this was a different kind of SamTastic Weekly Tip. It is also the longest Tip we’ve ever sent.
So, to bring this week’s Tip to an end, I offer two things:1. All SAM team members are eligible to receive The Main Idea. All you have to do is complete this online form: https://bit.ly/3ZVJkPx . We provide access at no charge to SAM team members.
2. You can read, below, the email I sent to Jenn David Lang, The Main Idea author, and the how I used AI to revise the message. Which version is best? Does AI capture my intent? I believe this illustrates the concern, and optimism, I have about AI and its future use in TimeTrack.
If you would like to share your use of AI, use this link: https://bit.ly/4fn9cbx
Hey Jenn.
Thanks for your work on the AI Playbook.
I appreciated the way you explained how AI could be used by an educator. The prompt examples were great.
I’ve wondered if a user of AI has an ethical responsibility to disclose its use in a communication or document. Was this question addressed?
We have a lot of principals who are using AI. A principal in Iowa drafts letters and newsletters and then asks AI to revise to make them more friendly and shorter. In this case, I am not sure disclosing the use of AI would be ethically required. If a principal did the reverse, ask AI to draft a newsletter covering five issues to share with parents in a friendly manner, it seems there may be a greater need to disclose AI use.
If you go a step further, and ask IA to revise, or write a feedback email, or evaluation, using notes the leader made during a classroom visit, or visits, then, it seems, there is a far greater responsibility to disclose.
Some principals tell me they use AI this way. I ask if they read and modify what AI creates before sending and disclose the use of AI. Most read. Few modify. None disclose the use of AI.
Full disclosure: We banned the use of AI by our Time Change Coaches last year until we can develop clear guidelines on use. At the same time, we are exploring how AI can be built into TimeTrack. With the narrative info principals enter for each teacher, project and student with whom they work…. coupled with the mathematical record of time spent, the potential use of AI is rich…and a bit worrisome.
Thanks again for your work, Jenn.
Mark
Mark Shellinger, Director
National SAM Innovation Project
The message above was written without AI assistance. The message, below, was ChatGPT’s response to the prompt: Would you rewrite this letter to Jenn and make it friendlier and more concise?
Subject: AI Playbook Insights
Hi Jenn,
Thanks for your excellent work on the AI Playbook. I especially appreciated how you explained how educators can use AI, and the prompt examples were very helpful.
One question I’ve been thinking about is whether there’s an ethical responsibility to disclose AI use in communications or documents. Did you touch on that in the playbook?
We have a lot of principals using AI, like one in Iowa who drafts letters and newsletters, then asks AI to make them friendlier and more concise. In this case, disclosure doesn’t seem as necessary. But if a principal asks AI to draft a newsletter on multiple topics or write feedback on classroom visits, it feels like disclosure would be more important.
I’ve heard from some principals using AI this way, and when I ask if they read and modify what AI creates, most say they do, but none disclose its use.
As a bit of background, we banned AI use by our Time Change Coaches last year while we work on clearer guidelines. We're also exploring how AI might be integrated into TimeTrack, especially with the narrative info principals enter about teachers, projects, and students. There’s a lot of potential, but also some concerns.
Thanks again for all your work, Jenn. I look forward to hearing your thoughts.
Best,
Mark
The AI version is more concise. It is 71 words shorter. It is also a bit friendlier, I think. Using AI this way took me more time, not less.
If, instead, I ask AI this prompt: “Write an email to Jenn, who wrote a summary of the book, AI Playbook, asking if a principal should disclose use of AI in a newsletter for parents and a feedback email or evaluation after observing a teacher. Here’s what I got:
Subject: Clarification on AI Disclosure in Principal Communications
Hi Jenn,
I hope you're doing well! I recently read your summary of AI Playbook and found it very insightful, especially the sections on how AI can be integrated into education. I have a quick question regarding the use of AI in school settings.
As a principal, I'm wondering whether it would be appropriate to disclose the use of AI in specific communications with parents, such as a school newsletter, or in feedback emails or evaluations after observing a teacher. Do you think it’s important to make parents and staff aware of AI tools being used in these contexts, or would it be more effective to keep it implicit?
I would really appreciate your thoughts on this, especially considering your insights from the book!
Thanks so much for your help.
Best regards,[Your Name][Your Position][Your School]
This AI version is even more concise. I would need to spend some time to edit to retain the original meaning and content, but this would take less time, not more.
Ready to try it? https://chatgpt.com/
Comments