🤖 Analyzing 36 state AI policy documents with NotebookLM

📺 Ditch Summit, a free virtual conference for educators

📣 Announcing: The Ditch Summit speaker lineup!

It's our 10th year for this free online conference for teachers.

Dates: Dec. 15, 2025, to Jan. 11, 2026.

FREE PD certificates, too!

More details coming soon …

📚 New AI resources this week

1️⃣ Helping K-12 schools navigate the complex world of AI (via MIT) — MIT Teaching Systems Lab is working with educators to collect stories and develop a non-prescriptive guide for AI integration in schools.

2️⃣ AI In Education Statistics By Usage, Adoption and Facts (2025) — A data-rich article showing how widely AI tools like ChatGPT, Grammarly, and Microsoft Copilot are being used by students and educators globally.

3️⃣ What Parents Need to Know About AI in the Classroom — A guide for parents highlighting AI’s uses in K-12, the questions to ask, and key issues around transparency and fairness.

📢 Your voice: AI literacy for students

Last week’s poll: What AI literacy concept is most important for students to learn?

🟨⬜️⬜️⬜️⬜️⬜️ How to do work efficiently with AI (7)
🟩🟩🟩🟩🟩🟩 How to maintain AI/human balance (39)
🟨⬜️⬜️⬜️⬜️⬜️ How to protect yourself from AI perils (9)
⬜️⬜️⬜️⬜️⬜️⬜️ How AI is changing the workforce / world (5)
⬜️⬜️⬜️⬜️⬜️⬜️ Something else ... (5)

How to work efficiently with AI: AI/Human balance is definitely important, but at this initial stage, I feel like our students need to know what AI is and what it can do to support their learning. — B. Keiser

How AI is changing the workforce / world: I do not think that education today is preparing students for the landscape of work in the future--and I think that change is occurring more quickly than we can possibly adjust. While there are crucial skills that will likely be relevant in the future--collaboration, critical-thinking, creativity, questioning--it feels as if we are tied to standards/assessments that are preparing students for a world that no longer exists. — R. Doty

How to maintain AI/human balance: I love the phrase you mentioned in an earlier newsletter this year; "The less important it is the more I lean on AI. The more important it is the less I lean on AI." I've come to use that as a general guideline for myself as well as in what i teach my teachers in PD sessions and my own newsletters. I think it pretty much sums up what's needed to be said here and it's an easy concept to grasp - regardless of who you are and what you do. And so, so important to understand. — Rickard, Sweden

How to protect yourself from AI perils: I think it is important for us to help students analyze AI as well as all media to determine validity, bias, and intent. So much of what we battle in education and the community at large is misinformation. — M. Rose

What would you like to read in AI for Admins?

What’s a topic you’d like to see covered here? Hit REPLY to this email and let me know.

Have you done anything you’d like to share with the AI for Admins community? Hit REPLY and let me know.

Would you like to write a guest post to support and equip AI for Admins readers? Hit REPLY and let me know.

🗳 Poll: State AI guidance

Instructions:

1. Please vote on this week’s poll. It just takes a click!

2. Optional: Explain your vote / provide context / add details in a comment afterward.

3. Optional: Include your name in your comment so I can credit you if I use your response. (I’ll try to pull names from email addresses. If you don’t want me to do that, please say so.)

🔎 Analyzing 36 state AI policy documents with NotebookLM

More than 30 states — and the territory of Puerto Rico — have published documents giving guidance to schools and districts about artificial intelligence. (I’ve also included the U.S. Office of EdTech guidance that has been subsequently removed from public availability.)

Some guidance documents have been out for a long while. Other states, like Alabama, Montana, Maine, Alaska, and Nevada, are newer.

Individually, they’re helpful documents to educators and school leaders in their respective states.

But collectively? When analyzed together as a whole, they give a glimpse into how the entire United States is guiding their schools.

About a year ago, I gathered all of the available state AI guidance documents (available here at AI for Education … grateful for their work to collect these!). Then I put them into a notebook in NotebookLM.

It allowed me to ask the AI questions about the collection of guidance documents as a whole … What’s common? What outliers are there? What’s missing?

Back then, I couldn’t easily share a NotebookLM notebook with you. But now I can. With this shared version, you can chat and ask questions about the documents — and access all of the audio, video, and text resources I’ve created in the Studio.

Here are some questions I’ve already asked it — with summarized, bullet point responses below.

What suggestions do these documents have for schools?

The documents suggest that schools and districts should address AI by focusing on the following actions:

• Prioritize a human-centered approach, ensuring AI augments human judgment and critical thinking, never replacing the essential role of educators.

• Develop comprehensive AI literacy training for all staff and students, addressing capabilities, ethical use, and potential biases.

• Update existing policies (AUP, academic integrity) rather than banning AI, ensuring clear governance and expectations.

• Advance academic integrity by redesigning assignments, requiring AI use disclosure and citation, and avoiding sole reliance on unreliable AI detection tools.

• Implement robust security measures, vet vendors, and explicitly prohibit users from entering Personally Identifiable Information (PII) into public-facing AI tools.

• Ensure equitable access, address algorithmic bias, and leverage AI to enhance operational efficiency.

What suggestions do they have about forming policy about artificial intelligence?

The following suggestions are offered for forming policy about artificial intelligence:

• Form an AI Governance Committee or task force with diverse stakeholders, including administrators, teachers, students, and community members.

• Review and update existing policies (Acceptable Use, data privacy, academic integrity) to explicitly address AI usage, rather than implementing immediate, outright bans.

• Ground policies in core principles such as human oversight, transparency, fairness, accountability, and adherence to data privacy laws (FERPA, COPPA).

• Define clear expectations for acceptable use, requiring students to disclose and properly cite when AI is utilized in their work.

• Establish a cyclical process for regular policy review, auditing, monitoring, and continuous adaptation due to rapidly evolving technology.

What suggestions do they have for strengthening AI literacy in schools?

The documents emphasize several suggestions for strengthening AI literacy in schools:

• Develop and implement comprehensive AI Literacy programs tailored for all stakeholders, including staff, students, and the community.

• Integrate AI literacy concepts (principles, ethics, bias, limitations, and responsible use) into curriculum across all grade levels and content areas.

• Provide differentiated professional learning and ongoing support for educators to build confidence and model ethical AI use effectively in instruction.

• Equip students with critical thinking skills to understand how AI works, recognize its limits, and evaluate outputs for potential inaccuracies and bias.

• Build AI literacy on existing digital citizenship, media literacy, and computer science standards.

Are there any outliers — anything that contradicts or doesn’t line up with other reports?

The most prominent outlier pertains to the use of AI detection tools:

• The Mississippi guidance suggests administrators should require students to submit work to plagiarism checkers and encourages educators to use AI detection tools. Rhode Island also includes AI detectors in its procurement checklist.

• This contrasts sharply with the nearly unanimous advice that schools should not rely solely on AI detection software due to its unreliability, high rate of false positives, and demonstrated bias against non-native English speakers and certain writing styles.

Other subtle outliers involve:

• High-Stakes Use: Rhode Island notes that AI can analyze historical disciplinary data to model bias-free consequences, potentially informing high-stakes governance, moving beyond the typical strict avoidance of AI in high-stakes decisions.

What gaps are there in these documents?

The following areas appear to be lightly addressed or represent potential gaps in the guidance documents:

• Explicit guidance and procurement policies addressing the environmental impact (e.g., energy consumption, carbon emissions) associated with training and running large AI models.

• Detailed strategies for mitigating student mental health and social-emotional risks stemming from potential over-reliance on AI companions or diminishing human interaction beyond cyberbullying/deepfakes.

• In-depth policy frameworks for managing bias and ensuring human oversight in high-stakes operational uses (such as HR, financial forecasting, and resource allocation) are generally lacking detail compared to instructional policies.

• Comprehensive policy addressing novel intellectual property rights and content ownership arising specifically from the use of generative AI outputs, often noted as an area of legal uncertainty.

• Specific focus on supporting the implementation challenges and technological needs of small, rural, or tribal districts (e.g., shared services and staffing models) is not universally detailed.

Ask your own questions about guidance documents in NotebookLM

Have a very specific question — or something I didn’t cover above — that you want to ask? Open the NotebookLM notebook and ask it yourself! And while you’re there, check out the audio, video, and text resources in the Studio.