Can Schools Trust AI to Translate for Non-English-Speaking Parents?
Special Report
Special Report
Artificial Intelligence

Can Schools Trust AI to Translate for Non-English-Speaking Parents?

By Alyson Klein — September 21, 2026 10 min read
Illustration by Nicole Xu showing a teacher in a teacher-parent conference with a laptop that translates a distorted depiction of the student. The female parent of the EL student has a swirly white line of confusion circling her head.
  • Save to favorites
  • Print

Middle school nurse Charlie Davis regularly contacts parents with urgent information: an injured child, missing vaccination records required to attend class, or an incomplete physical necessary for the soccer team.

Getting those crucial messages across can be tough when a caregiver isn’t fluent in English.

Davis’ district, Croton-Harmon in New York’s Hudson Valley, has turned to artificial intelligence to bridge what once seemed like insurmountable communication gaps, even as leaders there acknowledge it’s far from a perfect solution.

Platforms with translation capabilities must protect students’ private information, which is why the district carefully vetted the technology it is uses, said Ellen Moskowitz, the district’s director of technology.

And there are big questions about the accuracy of AI translations, she added.

No parent has complained that a message they received was “bonkers or didn’t make sense,” Moskowitz said.

When possible, the district checks translations with a native speaker. But that’s not always an option, since some students speak a language at home that no one on staff is fluent in.

“There is a little bit of blind faith that we’re putting into this,” Moskowitz said.

On balance, though, the translation platforms are an improvement over not being able to communicate clearly with many non-English speaking families, she said.

Nationally, about 1 in 5 children in the United States speaks a language other than English at home, according to the Century Foundation, a nonprofit organization.

Most of those children—about three-quarters—speak Spanish, the nonprofit reported. But tens of thousands of public school students speak other languages at home, including Arabic, Haitian Creole, Portuguese, and Russian, according to the National Center for Education Statistics.

In recent years, schools have turned to a variety of AI-powered tools to help translate materials into families’ home languages.

Educators might put text directly into a large language model, like ChatGPT or Gemini, and ask for a translation or use a system like Google Translate.

Apps that specialize in parent communication often include a translation component, though using these platforms districtwide—and in a way that protects student data— typically requires a paid subscription. Districts also use handheld translation devices and earbud translation tools to provide real-time translation during meetings.

Some of these tools predate generative AI but now incorporate the technology.

Educators see plenty of caveats along with the benefits.

“AI is a powerful tool, but any tool is only as good as the user,” said Sharon Mofield-Boswell, the director of family and community engagement for the Fayette County school district in Lexington, Ky., whose students speak more than 100 different languages at home. “AI can generate and translate content, but it can’t give context. It has to have some really clear boundaries, and it cannot take over the human component of translation and interpretation services.”

AI does some aspects of translation well, others not so much

Translation involves four key elements: the words, their basic meaning, the author’s intent, and the cultural context, said Belle Li, a researcher in Purdue University’s Learning Design and Technology program.

AI handles those first two elements “very well, even better than some professional translators,” Li said. But the technology often fails to accurately capture an author or speaker’s intent and the broader cultural context.

Even when its translations use the right words, AI doesn’t always reflect shades of meaning, Li said. For instance, if a teacher writes that they are “worried” about a student’s recent slow progress in reading, the technology might choose a word that better describes a dire situation, as opposed to a moderate concern.

AI might also misconstrue tone, striking a friendly, less formal approach than an educator intended. In the case of Spanish, that might mean using the more familiar “tú” form of a verb as opposed to the more respectful “usted.”

The technology also strives to resolve ambiguity, turning a word that means “might” into one that’s closer to “will.” Part of the reason: As the technology evolves, AI tools try to “improve” a text, not just translate it, Li said.

“The boundary between translation and rewriting is becoming blurry,” she said.

What’s more, some languages, such as French and French Creole, don’t use “a.m.” or “p.m.” when differentiating time—a nuance translation platforms may miss, said Heejung An, a professor of education leadership at William Paterson University in Wayne, N.J., who specializes in STEM and technology.

That can be problematic even when sending out a simple notice about the timing of a back-to-school night.

And AI isn’t adept at handling colloquialisms (think phrases like “It’s a piece of cake” or “You’re pulling my leg”).

For instance, if a teacher writes: “It would break my heart if anyone missed the Zoo field trip,” a platform might translate that literally, as though the teacher would go into heart failure if a student couldn’t join.

Educators wonder if it’s better to avoid sending a ‘gibberish’ document to families

AI’s translations also tend to be more accurate when the technology is working between two “high resource” languages—those that comprise a significant volume of what’s written on the internet, such as English, Chinese, Spanish, or French—as opposed to “lower resource” languages that don’t appear as often online, such as Swahili, Lao, or Yoruba, Li said.

But the technology can still struggle even between two high-resource languages, including English and Spanish, said Elizabeth Stringer-Nunley, the English-learner lead for the Galax City Public Schools district in rural Virginia and a 2026 Education Week Leader To Learn From.

She’s never been able to find an AI-powered translation technology that translates the Spanish word for “senior”—meaning 12th grader—correctly. Instead, the apps typically use the word for “senior citizen”—meaning older person. She’s also seen platforms translate “free” as in “no cost” to “free” as in “liberated.”

Perhaps the most groan-worthy mistranslation occurred during the COVID-19 pandemic: The Galax district wanted to encourage parents to stay out of school buildings, so staff created a sign saying: “Thank you for remaining in your car” and translated it into Spanish using Google Translate to “Gracias por dejar los restos en su vehículo.”

The literal meaning behind the sign the district put on sandwich boards outside the school? “Thank you for keeping the remains in your car.”

“Obviously, we didn’t want them to leave dead bodies in their cars, but that’s what it translated to because [the tool] used the noun form of the word instead of the verb form of remain,” Stringer-Nunley said.

In fact, Stringer-Nunley sometimes wonders whether it’s better to avoid machine translation altogether.

“I think it’s frustrating sometimes when [parents] get these documents that they don’t understand,” she said. “I think what is most appreciated is a real person trying and making mistakes [rather] than sending home a gibberish document.”

AI gives the ‘illusion of accuracy’

To be sure, AI has improved since the pandemic years, and particularly since the rise of generative AI.

Still, it might take a fluent bilingual speaker, such as Stringer-Nunley, to notice a problem, in part because AI-powered translation has become so quick and authoritative, Li said.

That can give educators the “illusion of accuracy,” Li said. “AI translation is so fluent it makes it harder to notice a miscommunication.”

Ideally, a bilingual speaker—whether a district-employed official translator or a school staff member who happens to speak another language—should review materials translated by AI before they are sent to parents, experts and educators say.

But that’s not always possible. There may be no adult at a school who speaks a given language—particularly one that’s less prevalent in the United States than Spanish, such as Urdu or Hausa.

To get around some of the cultural problems, Li recommends educators do more than just put text into a tool powered by ChatGPT or another large language model. They should let the tool know the context—the reason for the message, the tone they’re going for, and perhaps even mention that they are a teacher trying to communicate with a parent.

And educators can check the quality of one AI platform’s translation by putting it into another AI application and asking that tool to translate the message back into English, Li suggested.

Educators should also be transparent about using AI for translation, Li said. Many parents “know AI makes mistakes” and might appreciate the school’s effort to communicate with them in their language, despite the bumps.

Educators who use the technology frequently for written communication have their own advice.

“No euphemism. No slang. Be to the point. Be brief,” said Davis, the Croton-Harmon school nurse. “The more direct you can be, the better.”

And while educators are used to jargon-laden speech such as “We’re switching our diagnostic assessment from MAP to iReady,” that might sound confusing even to English-speaking parents, much less in translation, Mofield-Boswell of the Lexington, Ky., district said.

Not everything can or should be translated by AI

To be sure, there are some documents and communications—such as individualized education programs for students in special education or information on a student’s academic placement—that should never be translated by AI alone.

“I think the level of human oversight should depend on the stakes,” Li said.

The week’s lunch menu or the timing of school spirit week is a relatively low-stakes announcement where an AI translation is likely to be better than not providing the information at all, Li said.

However, school districts that send out sensitive information that has been incorrectly translated by AI may be opening themselves up to more than misunderstandings; they risk legal repercussions, too, said Amelia Vance, the founder and president of nonprofit advocacy organization Public Interest Privacy Center.

“You could be fired” for sharing a significantly inaccurate, high-stakes document with a family, she said. “It could lead to lawsuits.”

Privacy is a major consideration with AI translation tools

What’s more, unless a large language model or an app has an agreement with the district—which nearly always requires a paid subscription—student data aren’t secure.

Croton-Harmon has wrestled with these privacy issues.

When Davis was initially seeking a way to talk to non-English-speaking parents about their children’s medical needs, he signed up for a free, personal subscription to a popular school communication app that offers translation services.

But when he mentioned the platform to Moskowitz, she quickly pointed out that the app didn’t have a privacy agreement with the district. She steered him to a competitor that had already agreed to comply with Croton-Harmon’s student-data standards.

Since then, the district has expanded its use of that app and received overwhelmingly positive feedback from families.

The app doesn’t do everything, though, Davis said. He and Moskowitz still need to find a secure way for him to communicate in real time with families over the phone about an urgent medical situation, such as a student who fell and is being monitored for a concussion.

There are practical hurdles, too, even with the tools that Croton-Harmon already uses, Davis and Moskowitz said.

Parents must download the communication apps that offer translation services.

And teachers must switch from communicating with families primarily through email to contacting parents and caregivers on an app.

Those obstacles are worth traversing, Davis said.

“Communication with families is so vitally important to students’ success, their academic performance, their social-emotional growth and learning,” he said. “We have to use these tools, from an equity standpoint, to [give] our English-language-learning students and their families the same opportunity” as everyone else.

Coverage of English learners is supported by a grant from Charles and Lynn Schusterman Family Philanthropies, at www.schusterman.org. Education Week retains sole editorial control over the content of this coverage.

Events

This content is provided by our sponsor. It is not written by and does not necessarily reflect the views of Education Week's editorial staff.
Sponsor
School Climate & Safety Webinar
There But Not Present: Rethinking Student Engagement
Learn how to recognize the early signs of disengagement and what the behavioral science reveals about which interventions actually work.
Content provided by Frontline Education
Teacher Preparation Webinar What Teacher Education Should Look Like in the Age of AI
Explore the current trends around the use of AI in teacher preparation and continuing education with data from the EdWeek Research Center.
English Learners Webinar How Can School Districts Better Support English Learners With Disabilities?
Join this webinar to hear recent findings from the EdWeek Research Center and discover practical strategies for improvement.

EdWeek Top School Jobs

Teacher Jobs
Search over ten thousand teaching jobs nationwide — elementary, middle, high school and more.
View Jobs
Principal Jobs
Find hundreds of jobs for principals, assistant principals, and other school leadership roles.
View Jobs
Administrator Jobs
Over a thousand district-level jobs: superintendents, directors, more.
View Jobs
Support Staff Jobs
Search thousands of jobs, from paraprofessionals to counselors and more.
View Jobs

Read Next

Artificial Intelligence A Reckoning on Social Media Addiction Came Late for Kids. Will AI Safety Be Different?
Efforts to make social media safer offer lessons for those looking to protect children using AI.
5 min read
Kristin Bride, right, mother of 16-year-old Carson Bride, who died by suicide after being cyber-bullied on an anonymous app on Snapchat, is joined by Amy Neville, left, mother of 14-year-old Alexander Neville, who died from fentanyl poisoning after being deceived by a drug dealer operating on Snapchat, during a Social Media Victims Remembrance Day memorial on Capitol Hill, June 23, 2026, in Washington.
Kristin Bride, right, mother of 16-year-old Carson Bride, who died by suicide after being cyber-bullied on an anonymous app on Snapchat, is joined by Amy Neville, left, mother of 14-year-old Alexander Neville, who died from fentanyl poisoning after being deceived by a drug dealer operating on Snapchat, during a Social Media Victims Remembrance Day memorial on Capitol Hill, June 23, 2026, in Washington.
Rod Lamkey, Jr./AP
Artificial Intelligence Opinion Think Your Students Know More About AI Than You Do? Read This
Here are four things your students don’t understand about AI—and what you can do about it.
Sam Wineburg & Nadav Ziv
•
4 min read
102026 opinion Wineburg AI media literacy
Vanessa Solis/Education Week + Getty
Artificial Intelligence ChatGPT's 'Teen Mode' Is Unsafe for Students, Report Concludes
The chatbot failed to report unsafe interactions to parents in a timely manner, researchers found.
5 min read
Young man using a smartphone to ask an AI about exercise
Vanessa Solis/Education Week + Getty
Artificial Intelligence At OpenAI Forum, Discussion Focuses on Stronger Teen Safety Measures
Panelists put forward ideas on how to incentivize industry to develop responsible products.
4 min read
Washington, DC - October 5, 2026: Arthur C. Evans Jr., PhD, Chief Executive Officer of the American Psychological Association  speaks during the discussion. The OpenAI Forum hosted a panel on Responsible AI for Teens, how AI can support teens’ learning while promoting safe, healthy use. The program featured an OpenAI Forum panel with experts in learning, adolescent mental health, and family and online safety: Kate Ruane, Director of the Free Expression Project at the Center for Democracy & Technology, Chuck Melley, SVP of Policy, Advocacy and Communities at CodeAI, Arthur C. Evans Jr., PhD, Chief Executive Officer of the American Psychological Association and Mathilde Cerioli, Chief Scientist at everyoneAI. Chris Lehane, OpenAI’s Chief Global Affairs Officer, moderated the conversation and shared context on ChatGPT for Teens, a learning-first experience launched in August and designed to help teens think critically, deepen understanding, and use AI with confidence.
Arthur C. Evans, Jr., the CEO of the American Psychological Association, speaks on a panel at an Oct. 5, 2026, OpenAI forum on how to support responsible use of AI among teenagers.
Sarah Voisin for Education Week