Issue# 5: AI & Research at School

AI-generated cartoon of Jay McGrane sitting at a desk showing a laptop screen that says “AI and kids” while gesturing to a person off screen.

Written by Jay McGrane with an assist from Paula

Where is AI in today’s K-8 classroom?

Paula is taking a break this week so I thought you might like a peek behind the K-12 curtain. I sub in elementary classrooms twice a week, and let me tell you, AI hasn’t changed very much at all. 

Even though headlines are panicking about students using AI to write entire essays. In my experience, most of my students haven’t even tried AI…on purpose.

AI is just… there. 

Uninvited. 
Unannounced. 
Unremarked.
Sitting right at the top of Google.

Let me show you how AI has changed research for some of our youngest learners. 


It’s Not the Kids. It’s the Interface.

I call it the “Google and Paste.” 

Most kids type their question into Google and copy whatever it says. They don’t click any links. The source cited is “GOOGLE.” If they bother to cite at all… 

At this point, most op-eds tell you that kids are lazy.

(Sidenote: I might have been able to jump on this “lazy kids” bandwagon if my own beautiful daughter hadn’t come home doing exactly the same thing. And I can’t think of anyone less lazy.) 

No, kids aren’t “lazy.” UX designers are AMAZING at keeping people on the first page of Google. 

We’ve seen search evolve from links to snippets to popular questions to AI overviews. Google wants users to get their queries answered on the first page. So it isn’t surprising to me that’s how kids do research these days. (I’ll let you decide if looking up a quick question on Google even constitutes research…anyways that’s a question for another day.)


So…does AI actually change how kids research?

In my experience? It changes everything.

Marshall McLuhan was right: 

The medium is the message.

AI overviews present information with authority, even if the sources aren’t credible or don’t exist at all. Many summaries rely on subpar content like corporate blogs or Reddit threads. They also only include a generic icon to further obscure the credibility of a source. 

And, if kids don’t click the links, they miss the chance to assess bias or credibility. Most just see the link and think, “Cool, there’s a link. Must be right.”

AI blurs the line between authoritative and non-authorative sources in a way never before seen. Encyclopedias were always vetted by experts. Now, all that matters is if it sounds plausible.

Truthfully, kids are trying. 

I’ve had kids tell me that they “sleep” on the AI overviews. They read them one day and then write their presentation the next. Of course, they’re describing a great study skill without recognizing that it’s plagiarizing…and it’s not a research skill. 

So if copying from Google isn’t research, what is?


Meta-skills are the new basics 

To date, research skills in K-12 have focused on finding information. However, AI already finds information far faster than any human. 

The problem today has become accuracy and hallucinations.  Our skill sets need to change to work more effectively with AI systems. 

Research skills today need to focus on: 

  • Asking better questions

  • Judging credible sources

  • Summarizing for a purpose

  • Synthesizing to create something new

These skills allow students to be the human at the center. Even if they begin to work with an AI research buddy, they’re in control of the questions. They’re judging the source quality and attempting to synthesize something new. 

As we move into a world with AI agents, I think we have to continually put kids in the driver’s seat with AI. Instead of encouraging them to blindly copy the AI’s answer. 

This week, I’ll be watching how I model research at home. 

How often do I quickly look up something on my phone? 

Do I ever tell my daughter my process? 

Not yet, so I’m thinking about starting. I’ll keep you posted on how it goes.

What’s one meta-skill you wish someone had taught you sooner?

Hit reply. I’d love to add more skills to my list!

Jay 

P.S. Paula will be back for the next issue! 


What I’m doing as a parent in the AI chaos (Jay)… 

In response to “Google and Paste”, I’ve been kicking it old school…with some AI help. My daughter had some questions about Ancient Sumerians. Instead of going directly to Google, I asked DeepResearch to create a 500-word report for me on the Ancient Sumerians using credible sources. I also asked it to be at her reading level. Most elementary schoolkids can’t access the vast majority of information on the internet because they can’t read well enough. Then, we worked on pulling out the big ideas, essentially summarizing.

I know reading non-fiction isn’t particularly flashy, but I do believe it’s one of the greatest gifts we can give our kids. AI models generally give waaaay more information than required for even the most basic of tasks. 

The better your reading skills…the better your AI skills.

What tools I’m playing with (Paula)...

You didn’t think I’d be completely ignoring AI while I was out, did you?

I haven’t jumped into the new Google Gemini or Anthropic Claude updates yet, but I will. And Google’s new VEO3 video generator? Just… wow. I’ll have more to say about that soon. 

For now, I’m stress-testing AI for a new use case I think has real potential to keep humans at the center. More on that soon. Until then… the tools aren’t going anywhere. And neither am I. 😉


If you’re thinking more seriously about how to use AI in your work or classroom…

I’ve got classes coming. Human-centered. Accessible. Made for people like us. Click here to join the list.


We hope you’ve enjoyed our thoughts on how to keep the human at the center of AI. We’ll be back in 2 weeks with our next installment on AI and the future of work. Until then…

https://www.linkedin.com/in/paulamcconnell/
https://www.linkedin.com/in/jaymcgrane-edtechwriter/

Jay & Paula


P.S:
f this email gave you a moment of breath, relief, or a better question — there’s more.

The Clarity in the Noise is Seva Digital's free 3-part email mini-course that unpacks how to stay steady, smart, and human while everything accelerates.

It's not fluff. It's not fear. It's strategic calm delivered in plain language, with real tools. Start the Series Here.


AI disclosure: We use Riverside to record the conversation for the future podcast. Jay writes the newsletter (no AI), pulling quotes from the transcript with ChatGPT. Paula takes the final newsletter, adds her part with an AI assist, creates the image, and loads it into Flodesk. So sure, we use AI tools, but it is built on very human conversations and Jay’s excellent writing.

Next
Next

Issue# 4: Bias, where?