AI Hacks for Any Teacher Working with ELLs

Hey there, and welcome back to the Equipping ELLs podcast. My name is Beth Vaucher, and I am so grateful that you are joining me here today. Today, I wanna talk about something that most of you are probably using, experimenting with, or at least thinking about right now, maybe you’re afraid of it, maybe you’ve avoided it, and that is AI.

Specifically, I wanna talk about what AI can actually do for teachers that are working with multilingual learners. What it can do in a practical, useful way, and what it cannot do, which I think is just as important to understand, right? Because here is the honest truth about AI and ELL lesson planning. AI is a genuinely useful tool.

It saves time on certain decisions, it generates things fast, and there are specific ways to use it that will make supporting your multilingual learners easier. And this is really an episode that is for anybody who’s working with multilingual learners. So I wanna give you some of those first steps to take so that even if you’re like, “Language objectives? What are language objectives?” Or, “I hate writing language objectives.” How can we use AI to make those things easier for us and give our students that support? But let me say this, AI is not a curriculum. It is not a system, and it does not know your students. And if you’ve ever pasted a prompt into ChatGPT and gotten something that looked great on the surface, but did not quite work for your actual ELL students at their actual levels, you know exactly what I mean.

So today, what I wanna do is give you six specific ways that you can use AI as a teacher of ELLs that are genuinely helpful. Some hacks I’m gonna give you, okay? And after each one, I’m gonna be honest with you about where it stops being helpful and what you need beyond it. So let’s get into it. Now, before I give you the six strategies, I wanna offer you a framing that I think makes all of this click.

AI is really good at giving you a starting point, right? It’s fast, it can output things in seconds, and starting points are genuinely valuable when you are planning lessons or figuring out how to support your ELL students with limited time. But for ELL instruction, the starting point is not the hard part.

The hard part is everything that comes after it, right? Making it level appropriate, tiering it across groups, connecting it to a language function, building a speaking routine around it, adding visual supports, pairing it with an assessment that actually tells you something. That is the work, and AI does not do that work.

So think of AI as a shortcut to a starting point, and think of what comes after as where your expertise and the right resources take over. So again, in this episode, I’m not gonna walk you through which AI tool to use or how to set it up. This is not a how-to in terms of AI. What I wanna do is assume that you’re already at least familiar with one type of AI platform, and you’re looking for hacks on how you can more easily provide support for your multilingual learners, okay?

So these are just some hacks that you can throw into your lesson planning, throw into your support, and it’s gonna make things easier. It’s gonna help with some of those things that are time-consuming and can be difficult to do on our own. It’s gonna help speed up that process and help us be more consistent in doing those things.

So win number one, generate language objectives. How many of you love writing language objectives? Right. I don’t see any hands raised, right? Because this is one of the most immediately useful things you can do with AI, and one of the quickest wins for ELL planning, especially if you are a homeroom teacher.

Take your content standard, paste it into whatever AI you’re using, and ask it to write you a language objective for that standard at two or three proficiency levels. Something like, “Here’s the content standard I am teaching this week. Write me a language objective for a Level 2 ELL student and a language objective for a Level 4 ELL student using the language function of explanation.”

In about 15 seconds or less, you have something to work with, and for a task that used to take teachers 10 minutes, or that I would just avoid completely, it’s genuinely useful, right? But here is where that might stop. The language objective AI gives you is a starting point, not a finished product. We need humans using AI to verify, to check.

You still need to check it against your WIDA Can-Do Descriptors to make sure it actually is aligned to that proficiency level. It comes up with some crazy things sometimes, right? You still need to build the sentence frame that makes the objective achievable. You still need to make sure it connects to the vocabulary thread you’ve been building all unit.

So don’t think that one prompt and you’re gonna get the output you need and be ready to go. It does take some back and forth, some work, some checking to make sure that it is giving you something that is useful for your students. AI gives you a draft. That’s what I like to say. It’s a good first step.

You give it the context that makes it right for your specific students. Now, in the next few weeks, we are going to be going deeper on WIDA specifically, so you will know exactly how to check that alignment and make those decisions with confidence. But to me, creating language objectives, especially if you’re a homeroom teacher, use it for that.

Take those content standards, whatever subject, those math standards, your reading standards, and create language objectives. That’s gonna help you guide your lesson when you have a solid language objective. Win number two, build vocabulary lists and student-friendly definitions. Here you can paste your text or your content topic into AI and ask it to pull the 10 most important vocabulary words for ELL students, along with student-friendly definitions at a specific language level.

Now, maybe you don’t wanna do 10, maybe do six. It’s up to you. You know your students. But this can really genuinely be a time-saver. So instead of reading through a text and manually identifying tier two and tier three words and then writing definitions that are actually accessible to a level two student, AI can draft that list in seconds.

You can even ask it to write the definition the way you would explain it to a student verbally. Simple sentence structure, no jargon, an example built in, right? You give it those words exactly, and because you’re being very specific in your prompt, it will make a big difference in what you get back. But here is the honest limitation.

A list of words with definitions is not vocabulary instruction. Vocabulary instruction is repeated exposure in speaking, reading, and writing across multiple days until those words are truly acquired. AI gives you the definition, which is a huge win, but it doesn’t give you the picture card, the speaking prompt, the sentence frame, and the close reading passage where those words show up again in context.

All of those things together are what actually moves vocabulary from a list into a student’s language. But again, having a list done for you with definitions and examples that are appropriate for the language levels that you’re working with, that is a win. So if that’s all you do to scaffold for your students at first, that’s great.

Start there. You can always add on as the year goes on. Win number three, generate comprehension questions at different levels. Here you can take any text your students are reading and paste it into AI. You can then ask for comprehension questions at two or three language levels, right? Basic recall for level one and two students, inference for level three, evaluation or analysis for level four.

Again, super fast, very useful, and genuinely saves you time compared to writing every question yourself. And you can push it further. You can ask AI to write the comprehension question and a sentence frame for the student’s response. So level two students get the question and a frame, “I think blank because blank.”

Level four students get the question and a more complex sentence starter. That combination is actually quite useful, and it significantly reduces your prep time, right? Now you’re scaffolding for your students quickly and easily. But here’s where AI runs out of road. The questions are not formatted for your students.

They are not attached to a visual that makes the text accessible. They do not come with the vocabulary pre-teaching that level one and two students need before they even attempt comprehension, and they are not organized in the kind of scaffolded close reading experience that produces language growth.

So you do have the questions, which are super helpful, but you wanna make sure that you’re still building the lesson around them in ways that really help your students acquire the language. Win number four, simplify a complex text. This is great, especially for your older ELL students. This is one teachers reach for constantly, and honestly, AI genuinely does well with it.

You can paste in a grade-level passage, maybe it’s a science text or a social studies article or a novel excerpt, and you can ask AI to rewrite it at a specific reading level or for a specific language proficiency. Simpler sentences, shorter paragraphs, high-frequency vocabulary where possible.

Again, this can be done quickly, and it’s often pretty accurate. But here is the distinction I want you to hold onto, and this one matters a lot for ELL teaching. Simplified is not the same as scaffolded. A simplified text reduces the language complexity. A scaffolded text makes the original language accessible, and for ELL students, the goal is almost always scaffolded access to grade-level text, not a watered-down version of it, because students cannot acquire academic language they are never exposed to.

AI can simplify, but it cannot scaffold. It does not add the visual supports, the vocabulary previews, the annotated margins, the comprehension frames that turn a hard text into an accessible one. Simplified text is sometimes what you need, but make sure you know the difference and you are reaching for the right tool.

So there is absolutely a time and a place where simplifying a text is appropriate. It is helpful. Maybe that’s the first step of your scaffold, is to simplify it and help them to understand the gist of the grade-level reading, and then you build on as you go with that reading by adding in more scaffolded support so that by the end they are reading and comprehending the grade-level passage.

That’s your end goal. So there is a time and a place, and AI helps simplify text in seconds.

Win number five, draft a parent communication in another language. This one is super useful, and I find sometimes it’s not our first resort when we go to translation. A lot of times we’re used to going to Google Translate or something like that, but AI actually can do this really well. So you can type the message that you want to send home, maybe it’s a welcome note, an update on how a student is doing, information about a conference, and then you can ask AI to translate it into the language of your students.

Okay? It’s fast, it’s accessible, and for families who never receive communication in their home language, it is a meaningful gesture that you took the time and that you’re making the effort so that they can understand what’s going on in their native language. Now, AI gives you translated words, but as we talked about in episode 220, parent communication with EL families is about so much more than translated words, right?

It is about the personal connection that a correctly translated but impersonal paragraph does not create.

Because the translation, again here, that’s the easy part, but the relationship is the goal.

All right, win number six, create tiered sentence frames. Here you can give AI your language function and your proficiency level and ask it to write a sentence frame. Something like, “Write me a sentence frame for the language function of comparison at a level two proficiency level.” Then ask for level three, and then for level four.

Okay? Again, this is such a fast output, and if you’ve been listening to this podcast all summer, you know how central sentence frames are to everything: speaking, writing, reading, comprehension responses. So having a quick way to generate them is a great way to be scaffolding constantly inside your classroom with what you’re teaching.

All right, here is the honest truth about AI and supporting ELLs. This is what I want you to take away today. AI saves you time on the small decisions. Things like language objective drafts, vocabulary definitions, comprehension questions, simplified text, right? Sentence frames, parent communication translations.

These are real time savings, and I want you to use them. Figure this out: what’s one way that you can implement AI this week that’s gonna help you break down some of these more, I’m not gonna say annoying tasks, but more mundane tasks, or things that we know are really useful and helpful, but maybe we’re just not great at doing them on our own.

AI can do that for you. But ELL instruction is not a collection of small decisions, right? It is a system. A vocabulary thread that runs through speaking and reading and writing for four weeks. Tiered scaffolds that are organized by language function and ready to adjust up or down based on what you observe.

Close reading passages that are pre-scaffolded at every level and ready to pull for any topic. Assessment tools that connect directly to WIDA levels and tell you exactly what a student needs next. AI gives you pieces, but a system gives you a classroom that runs. This is what the Equipping ELLs membership exists for.

Everything we talked about today, you can find inside with our ready-made templates, our resources that are already done for you. It’s all there, it’s organized, and it’s been tested with real ELL students, ready to pull on Monday morning. So, use AI for the small pieces. We even have AI tools inside of our membership.

We have one where you can take a picture of grade-level content, and it will scaffold it out for you. It will pull out the vocabulary words to pre-teach. It will give you different suggestions of how to scaffold. So those are just small ways that we have built into our platform so that we can help you save time, use AI, and also have all the things already done for you, so that you don’t have to piece together or spend lots of time going back and forth with AI trying to figure out how to build out a lesson for your students, right?

We would love for you to come and try the membership out for one week for free. No credit card required, no commitment. We just wanna give you full access to everything that we talked about today so you can see if Equipping ELLs is a great fit for you this upcoming school year. You can go to equippingells.com/trial to start your free week right now.

Thank you so much for being here today. AI is a real tool. It’s not going away. It is there. It can be helpful, and I hope that these six strategies save you real time this week. Use it for those small decisions. Save your energy for the things that require your expertise, things like knowing your students, observing what they need, and building the kind of instruction that actually moves them forward.

If you’ve enjoyed these episodes, I would love for you to subscribe wherever you listen.

And if you have even just one minute to leave a quick review, it means so much to us. It helps other educators just like you come across this podcast and also take a listen and learn how to better support their multilingual learners. So, keep equipping your English language learners. I’ll see you next week.