Issue 35 • May 7, 2025

I’m always finding new ways AI can help me.

Just this past week, it figured out group rotations for class presentations. I gave ChatGPT the list of teams, available times, and number of tables, and it built a complete presentation schedule. After a few quick tweaks, I copied the chart into a document and printed it out.

AI also acted as an extra set of eyes when I was preparing a vinyl logo to iron onto a shirt. I uploaded a photo of the shirt with the logo stuck in place. ChatGPT told me the right side needed to move down a few millimeters. I adjusted, uploaded another photo, and ChatGPT said it looked great. That gave me the confidence to go ahead and iron it on.

Even as an experienced AI user, I keep discovering surprising ways it can help.

Chat Comebacks 🔗

I start AI chats for all kinds of things, and I love that I can pick up where I left off—even if it’s been months. But finding old chats can be tricky. That’s why I make a habit of renaming important ones. Most chatbots, like ChatGPT and Gemini, let you name chats, and Gemini even lets you pin them to the top of your sidebar for easy access.

I’ve also developed a simple workflow that helps me jump back into the right chat fast. With the chat open, I copy its unique URL from the address bar and paste it into my notes or the document I’m working on. That way, I can revisit the conversation exactly when and where I need it.

Even though each chat has a shareable URL, only I can access it—since I have to be logged into my account.

After hundreds of sessions, this little system helps me keep track of the ones I want to return to, whether for reference or to keep the conversation going.

Manners and Machines 😇

I struggle with whether I should be polite to an AI chatbot. On one hand, it’s not human and doesn’t have feelings. On the other hand, I am human, and being polite is part of who I am.

Some argue that polite language can actually improve AI responses. It puts the human in a conversational mindset, which often leads to clearer prompts. Plus, if the AI was trained on respectful and collaborative interactions, it may mirror that tone back.

Last month, OpenAI CEO Sam Altman replied to a social media post questioning how much electricity is used when people type “please” and “thank you” to ChatGPT. His answer? “Tens of millions of dollars.”

After reading that, I’ve stopped thanking AI after it replies. But I still write my prompts politely. As a human, it just doesn’t feel right to be rude—even to a machine.

Math Missteps with Machines 🧮

AI chatbots use Large Language Models (LLMs) to generate responses by predicting plausible word sequences based on the input, often with a bit of randomness to keep the language sounding natural.

While it might seem like chatbots should be great at math, LLMs are not inherently reliable for calculations. They don't “do math” like a calculator or spreadsheet. Instead, they generate answers based on patterns found in their training data. This means they can give incorrect results—sometimes with confidence—even if the formatting looks convincing.

I was reminded of this when I double-checked the sum of these numbers:
2, 2, 2, 2.5, 1, 3, 2, 2, 2.75, 4, 1, 1.5, 1, 5, 1, 2.5, 6, 1, 2.5, 5, 4

I pasted the list into ChatGPT and asked it to add them up. It confidently replied that the sum was 52.25. When I asked, “Are you sure?” it insisted that it was. But my own math didn’t match.

To get a more accurate result, I asked the chatbot to write code to calculate the sum. Once the code was run, ChatGPT admitted its earlier answer was wrong. The correct total is 53.75.

Next time I need to do math, I’ll stick with a spreadsheet. And if I do ask a chatbot, I’ll have it solve the problem with code instead of relying on a predicted answer.

Build with Blocks 🧱

Have you noticed the new panel in Google Slides? It appears on the right side of the screen and offers quick access to tools for inserting images and adding slides from Google’s template gallery.

The most exciting feature in this panel is Building Blocks. A building block is a pre-formatted piece of content you can easily insert and customize. Instead of manually arranging text boxes, tables, or shapes, you can drop in a ready-made layout like an Agenda, then tweak the text, colors, fonts, and layout to fit your needs. (Tip: Try ungrouping the elements to make editing even easier.)

But agendas are just the beginning. You’ll also find building blocks for Lists, Key Statistics, Quotes, and Headlines. Simply choose one, add it to your slide, and make it your own.

Dig a little deeper, and you’ll discover categories like Badges, Collaboration, and Phrases. These blocks work well as digital stickers to call attention to important content. Teachers might even use them to give quick, visual feedback on student slides.

Once you start using building blocks, you’ll wonder how you ever lived without them!

Curious Combinations 😳

Sparky is a new Google Arts & Culture experiment. It’s a playful online activity designed to spark creative thinking and connect students with inventors from the past whose ideas have shaped our daily lives.

To get started, students choose a prompt category:

• Invent a gadget to help eat your favorite food

• Create a new instrument to make music

• Design a vehicle for travel

Next, they drag two objects from a list on the left onto a digital sheet of paper and describe what the combination is designed to do.

AI generates a visual prototype and description based on the idea. It also links the student’s creation to a real invention from history.

Sparky is a quick, low-prep activity that doesn’t require logins. It’s a delicious digital snack, and I’ve added it to the Cognitive section of my Digital Snacks collection of brain breaks.

Personalized Productions 🎬

Don't spend your time making an end-of-year video for your students. Instead, give them the media and let them create their own personalized movies.

When I last taught 5th grade, I shared photos and videos taken throughout the school year in a Google Drive folder. My fifth graders picked and chose the media they wanted to include. At the time, they used WeVideo, but now they could use Canva, Adobe Express, or Google Vids. Each student ended up with a video full of their own memories.

Creating their own videos gave my students a chance to look back on the year, highlight what stood out to them, and tell the story in their own way. It was a fun and meaningful way to reflect on their time in fifth grade.

The bell is about to ring on this issue of Learning in Hand! If you found something worth highlighting, send it to a colleague. Extra credit if they subscribe. 😉

Smiles from Iowa,

Subscribe at learninginhand.com

Al was used to revise and proofread some of the text in this newsletter. It was also used to generate possibilities for section titles.