Paul Beckermann 0:01 Welcome to Tech Talk for teachers. I'm your host, Paul Beckermann.

Transition Music with Rena's Children 0:05 Check it out. Check it out. Check it out. What's in the toolkit? Check it out.

Paul Beckermann 0:16 The topic of today's episode is improving classroom AI prompts. By now, nearly everyone has at least experimented with generative AI chat bots. Tools ChatGPT, Google Gemini, Microsoft Copilot, or Anthropic Claude, they seem pretty intuitive to use. You type in a question or ask something to be created and hit enter. A few seconds later, the chat bot spits out an answer. It's quick and easy, but still, the results are not always consistently great.

Sometimes they're clear winners, and other times we're left shaking our heads and asking, "How did it come up with that?" Most times, the difference in output quality is based on the quality of the input prompt. The old saying, "Junk in equals junk out," there's a lot of truth to that with AI prompting as well. So in today's episode, I'm going to share three tips to help you get better results from your prompts.

Paul Beckermann 1:15 Here are your three. Here are your three. Here are your three tips.

Paul Beckermann 1:20 Tip number one is to have the proper mindset. I find it really helpful to think about interacting with an AI chat bot as if it were a trusted colleague. These chat bots are designed to be good communicators. They're trained on language, so we can interact with them conversationally. They're also very smart. They've been trained on vast amounts of knowledge, so we can ask some questions and they may have an answer. They're also really good at making connections.

We can give them information or lists of ideas and ask them to make sense out of it. Maybe we want to see trends in a series of data sets or find common strategies from a collection of educational articles. They're really good at connecting dots. However, we do need to remember that, just our colleagues, chat bots can also be wrong, or they could pitch us ideas that we just don't. That's okay. It's just any brainstorming experience.

It's important to remember that you are in the lead role when engaging with AI chat bots. They're great collaborative partners, rich sources of information, wells of inspiration and ideas. But in the end, it's up to us to process what we're given and make the final judgments. Oftentimes that means taking what you while making the necessary revisions and additions to make the outputs usable.

Tip number two is to write effective prompts. Now this is another place where it's helpful to think of our interactions with AI as interactions with colleagues. If we're not clear in what we ask or say, the person on the listening end won't give us great feedback in return. If I ask a friend to bake me a cake, I'll probably get a cake, but I'm leaving a lot to chance. When do I need it? How big should it be? What flavor, for what occasion? What's the budget? Does it have to be refrigerated? Without those details, there's a good chance the cake will not meet my needs.

The same goes for prompting. If I simply ask a chat bot to create a lesson about ecosystems, the chat bot won't know what the grade level is, the specific standards I'm targeting, the length of time I have to teach it, the technology I have available, how many students are in my class, whether this is an introductory unit or an advanced lesson. Again, there are too many unanswered questions.

A much better prompt might be: "Design a three-day inquiry-based science lesson on ecosystems for fifth graders that integrates digital resource tools and ends with a student-created multimedia project. Include learning objectives, vocabulary, and a formative check each day. Generate this as a downloadable Word document". This more detailed prompt will almost assuredly leave you more satisfied with the lesson than the first prompt. There are lots of prompting models available to help guide you. I'm going to share my list.

It's sort of a greatest hits of what's out there, and it's a system that I've had success with. Here are five components that can lead you to prompt writing success:

Number one, assign a role. Tell the chat bot what role they should assume. Most commonly, this means saying something act as an expert K-12 instructional coach, or you are an experienced ninth-grade life science teacher. Assigning a role doesn't magically make the AI into that person, but it does help them to know what data resources to pull from. It focuses its energy and gives you better quality results. Instead of searching the entirety of its knowledge base, it will focus on content specifically relevant to those roles and needs.

Number two, describe a clear task. Ask. Now this one seldom gets forgotten, but you do need to tell the chat bot what you want it to do. Do you want it to: design a lesson, create a rubric, generate an exit ticket? It's helpful to use action words design, explain, generate, analyze, and produce.

Number three, identify the audience. This is an important question, especially if you're using AI to help design lessons. While you should never input specific student names or identifiable data attached to students into the chat bot, you should describe in general who the lesson is for. What's the grade, what class, what's the overall achievement level, what are general interest areas? Are there specific things you're targeting? If it's not for students, describe that intended audience instead.

Number four, specify the final product. Tell the chat bot what you want the output to look. What kind of product is it? What's the format? Is it a bulleted list of ideas? Is it a rubric in table format? Do you want a process described in a step-by-step list? Maybe you want a script or an email or something else? The more clear you are about what you want, the better chance that you will get it.

And number five, add relevant limiters and context. This is where you use your judgment and ask yourself, what else should my virtual friend know about this task that I'm asking it to complete? It's the part of the equation that can really advance you to the next level and give you much better results. What additional information might improve the final product that you want to receive? This might include asking the chat bot to produce content at a certain reading level, or perhaps in a certain style. Is the intent to entertain, or should it be more serious? Do you want it to be skimmable, or something that may require more in-depth study? Are there length limits or specific examples you want included? Essentially, what are the relevant details you'd want your collaborative colleague to know if they were helping you complete this task at a high level?

Tip number three is to follow up and refine. Even with your best efforts, it's hard to get perfect outputs from your first prompt attempt. AI will inevitably pull something you hadn't expected. Perhaps you forgot to mention something that is obvious. Now when reviewing the results, therefore, it's really important to follow up and refine. This means asking follow-up questions or giving additional directions. Maybe the output was too wordy or complicated. You could ask for the response to be simplified. A classic example is to explain it to me, I'm a fifth grader. You can ask for the format to be refined. If you the fourth idea of a brainstorm list the best, you could say, Number four is the best. Generate more that.

Sometimes the AI will ask you if you want certain refinements. That's really helpful and might steer you in directions that you hadn't considered. It might ask you if you want something formatted in a different way, or if you want something expanded or clarified. Again, it's a colleague asking helpful questions. In general, keep the back and forth going until you're satisfied. It's very likely you'll spend more time refining than prompting, and that's okay. Although you might get tired, the AI won't.

Transition Music with Rena's Children 8:22 Here's a bonus tip.

Paul Beckermann 8:24 While these three tips will help get you better results, it still takes practice. The best way to get better at prompting is to do it often, review your results, and learn from them. I'd suggest picking a teaching task that you want help with. Maybe it's generating journal topics for a writing class, or developing engaging ways to develop community in your classroom, or maybe it's finding new exit ticket ideas. Whatever it is, pick one and spend a week or two using generative AI as a thought partner in brainstorming ideas. And do it on a regular basis. You'll quickly learn what works best for you, and using the tips in this episode is a good place to start. Oh, yeah. And if your students are allowed to use generative AI, definitely share these tips with them and have them level up their prompting as well.

Paul Beckermann 9:15 To learn more about today's topic and explore other free resources, visit AVID Open Access.org. Specifically, I encourage you to check out the article collection, AI in the K-12 Classroom. And, of course, be sure to join Rena, Winston, and me every Wednesday for our full-length podcast, Unpacking Education, where we're joined by exceptional guests and explore education topics that are important to you. Thanks for listening. Take care, and thanks for all you do. You make a difference.