Estimated reading time: 8 minutes
The best AI tools for English Learners are not a single product but a small stack, one tool to create materials, one to build reading and vocabulary, and one to give every student speaking practice. This guide groups the leading options by category, gives district leaders a way to evaluate them for standards, privacy, and evidence, and explains how to fund them. It is organized by job to be done, not by brand.
Table of contents
How to Evaluate AI Tools for English Learners
Before comparing products, agree on criteria. The strongest tools for English Learners score well on:
- Standards alignment: does it map to WIDA, TELPAS, ELPAC, or your state’s ELD standards?
- Data privacy: is it FERPA and COPPA compliant, with a clear data-processing agreement?
- Differentiation: can it adapt to proficiency levels, not just grade?
- Speaking practice: does it build oral language, the domain ELs get the least of?
- Evidence and support: is there research and district implementation support?
- Funding fit: is it an allowable, supplemental Title III purchase?
The Best AI Tools for English Learners, by Category
Most districts combine one tool from two or three of these categories.
| Category | What it does | Examples |
|---|---|---|
| Material generation | Create and differentiate lessons, worksheets, and assessments from prompts | ChatGPT, Claude, Gemini, MagicSchool, SchoolAI |
| Adaptive reading & literacy | Personalized reading, phonics, and vocabulary practice | Amira Learning, Lexia, IXL |
| Speaking & oral language | Individual conversational practice and speaking feedback | Telo AI |
| Access & translation | Simplify texts, translate, and support families | General assistants, translation tools |
Material Generation (ChatGPT, Claude, Gemini, and school-specific tools)
General assistants like ChatGPT, Claude, and Gemini are the most flexible and lowest-cost way to generate standards-aligned materials, when teachers know how to prompt them. School-specific tools like MagicSchool and SchoolAI wrap the same models in teacher-friendly templates and add school data controls. Choose the general assistants for flexibility and cost; choose a school-specific tool for guardrails and ease of adoption. To generate your own materials well, see our AI prompts for ELL teachers.
Adaptive Reading and Literacy (Amira, Lexia, IXL)
These platforms personalize reading, phonics, and vocabulary practice and are widely used with English Learners. They are strong on literacy and progress data. Their limit for ELs is the same one classrooms have: they build receptive and reading skills well, but little productive speaking.
Speaking and Oral Language (Telo AI)
Speaking is the domain English Learners get the least of and the one most likely to hold back a proficiency score, and it is the gap the categories above leave open. Telo AI is built for exactly this: it gives every student individualized conversational practice, in English plus Spanish and French for bilingual programs, aligned to WIDA, TELPAS, and ELPAC, with teacher analytics on participation and growth. It is available on the web and on an embodied AI robot. Telo complements, rather than replaces, your material-generation and literacy tools: they build the lesson and the reading; Telo scales the talking. A dedicated speaking tool is a focused kind of AI tutor, aimed at oral language rather than academic content. Book a demo to see it with your students.
What to Prioritize by Role
- Teachers and coaches: start with a material-generation tool you can prompt well; it saves the most time immediately.
- EL/Multilingual directors: weigh standards alignment and the speaking gap, since oral language is where programs most often fall short.
- Curriculum and technology leaders: lead with privacy (FERPA/COPPA), interoperability, and evidence.
How to Fund AI Tools for English Learners
Most AI tools for ELs are allowable, supplemental purchases under Title III, as long as they add to rather than replace core services. Districts often buy faster through cooperative contracts like TIPS, which can avoid a separate RFP. Confirm the supplement-not-supplant rule with your federal programs office before purchasing.
Building the Stack Without Overlap
The goal is a small, complementary set of tools, not a drawer full of overlapping subscriptions. Because the categories do different jobs, they combine cleanly: a material-generation tool helps the teacher prepare, an adaptive literacy platform builds reading, and a speaking tool scales oral practice. Problems arise when a district buys two tools that do the same job, two material generators, or two literacy platforms, and pays twice for one function while leaving another gap open. Map your current tools against the categories before buying anything, and you will usually find one or two categories already covered and one conspicuously empty. For most EL programs, the empty one is speaking, because classrooms and reading platforms both leave oral practice thin.
Sequencing the purchases also helps. A sensible order is to start with the material-generation tool, since it saves teacher time immediately and costs little, then confirm your literacy coverage, then add the speaking tool that closes the oral language gap. Buying in that order lets each tool prove its value before the next is added, and it keeps the total stack lean. A stack of three well-chosen tools that each own a distinct job will outperform a pile of six that overlap, and it is far easier to fund, train for, and defend to a school board.
The AI Policy Layer for EL Tools
Any AI tool that touches student data now sits inside a policy environment that has expanded quickly. As of 2026, roughly 33 to 35 states plus Puerto Rico publish official K-12 AI guidance, and a growing number of districts have board-adopted AI acceptable-use policies; two states, Ohio and Tennessee, legally require districts to have an AI policy. For EL tools this has practical weight. Every tool in the stack should have a signed data-privacy agreement and demonstrate FERPA and, for under-13 students, COPPA compliance. Speaking tools deserve particular attention, because they process a student’s voice, which the 2025 COPPA amendments treat as covered personal information, so ask specifically how voice data is used, retained, and deleted. Checking each tool against your state’s guidance and your own district policy before purchase is no longer optional diligence; it is how a district stays compliant while adopting the tools its English Learners need. Tools built for K-12 from the ground up tend to make this alignment straightforward rather than a retrofit.
Frequently Asked Questions
What are the best AI tools for English Learners?
There is no single best tool; most districts combine a material-generation tool (ChatGPT, Claude, Gemini, or a school-specific wrapper), an adaptive reading platform (Amira, Lexia, or IXL), and a speaking-practice tool (Telo AI) for the oral language gap.
What are the best AI tools for teachers of ELs?
For creating materials, the general assistants are the most flexible, and school-specific tools add guardrails. Pair them with a dedicated speaking tool so students get the oral practice materials cannot provide.
Are AI tools for English Learners safe for student data?
They can be, if the vendor is FERPA and COPPA compliant and signs a data-processing agreement. Keep student data out of consumer AI tools and follow your district’s AI policy.
Can we use Title III to buy AI tools?
Usually yes, as a supplemental purchase that adds to core services. Confirm with your federal programs office and consider cooperative purchasing to move faster.
How do we avoid buying overlapping AI tools?
Map your current tools against the categories, material generation, adaptive literacy, and speaking, before buying. You will usually find one or two categories covered and one empty, often speaking. Buy to fill the gap, not to duplicate a job you already pay for.
In what order should we buy these tools?
Start with a material-generation tool, which saves teacher time immediately and costs little, then confirm your literacy coverage, then add a speaking tool to close the oral language gap. Buying in that order lets each tool prove its value before the next is added and keeps the stack lean.
What should we ask about a speaking tool’s data practices?
Because speaking tools process a student’s voice, which the 2025 COPPA amendments treat as covered personal information, ask specifically how voice data is used, retained, and deleted, and confirm a signed data-privacy agreement plus FERPA and COPPA compliance and fit with your district AI policy.
Related Resources
- AI for English Learners
- AI Prompts for ELL Teachers
- Can Title III Pay for AI Tools?
- AI Tools for Schools
Conclusion
The best AI tools for English Learners work as a stack, not a single purchase: one tool to create materials, one to build reading, and one to give every student speaking practice. Evaluate each on standards, privacy, differentiation, and evidence, fund them through Title III where allowable, and make sure the stack closes the oral language gap that classrooms alone cannot.
Sources: U.S. Department of Education, NCELA; WIDA Consortium; U.S. Department of Education, Student Privacy.
The One Job a Tool Stack Still Has to Cover
Assemble the strongest possible stack and a familiar pattern emerges. The material generator produces beautiful lessons. The literacy platform tracks reading growth. Both are genuine gains for a teacher stretched across a room of English Learners. Yet neither can do the thing an English Learner most needs to do, which is talk.
Speaking is where proficiency is won or lost, and it is the domain a full class squeezes hardest. Documents and reading platforms build receptive skill efficiently; they cannot listen to a student, respond, and adjust. That limit is about capacity, not about the quality of any tool in the stack.
This is why the districts that evaluate carefully treat oral language as its own line item rather than an afterthought. Telo AI is one example of a tool built for that gap, giving each student individualized conversational practice aligned to WIDA, TELPAS, and ELPAC, in English plus Spanish and French for bilingual programs.
See how a speaking tool fits alongside material and literacy platforms: https://mytelo.ai/how-telo-works/
A well-chosen stack removes almost every constraint except the oldest one, which is that talking takes a partner. The districts that get the most from AI buy for the whole job, then make sure the part no worksheet can cover is actually covered.