AI Tutors Now Talk to Kids—But Who’s Watching the Screens?
Screen-sharing AI tutors promise interactive lessons for kids, but raise big questions about privacy, content quality, and parental oversight.
AI Tutors Are Watching—and Talking—to Kids Learning Code
Fifth- and sixth-graders are now getting help from AI tutors that can see their computer screens and talk to them in real time, according to a new study published on November 16, 2026, on ArXiv. The research looked at how children interact with "Gemini," a multimodal AI assistant, while solving programming challenges in Scratch—a popular coding platform for kids. The findings could reshape how millions of children learn, but they also spark new worries about privacy, content quality, and the role of adults in the classroom.
What Happened
The study, titled "Share your Screen and Talk to Gemini," explored what happens when young learners use an AI tutor that can both see their screens and respond to their voices. Instead of reading long, text-based hints, students could ask questions out loud and get spoken feedback. The AI could also "watch" their progress on Scratch projects, offering advice based on what was happening on screen.
This approach aims to make learning programming less frustrating and more interactive. Traditional AI chatbots often struggle to help kids who get stuck on visual tasks, like debugging code blocks in Scratch. By combining screen sharing and voice, Gemini could give more relevant, timely guidance. But the system also means children are sharing more of what they do—and say—online with an AI that’s always watching.
While the study focused on a classroom setting, the technology is part of a much wider push by companies like Google (which recently invested over $300 million in AI education tools) and OpenAI to bring AI tutors into schools and homes. Yet, there’s little regulation in the U.S. specifically covering how AI tutors handle kids’ screen data or conversations, beyond the Children’s Online Privacy Protection Act (COPPA), which was written long before today’s AI advances.
How This Affects Everyday People
For parents, teachers, and students, this technology is already knocking on the door. If your child is learning to code—or even just doing homework—there’s a good chance they’ll be offered an AI helper that can see their screen and talk to them. This could mean faster answers and more personalized help. For example, a child stuck on a tricky Scratch bug might get a friendly, spoken hint that helps them move forward, rather than a confusing wall of text.
But there are trade-offs. Screen-sharing means the AI can "see" everything your child is doing on their computer. That raises questions: What if the AI misinterprets what’s happening? What if it stores or shares sensitive information? And how do parents know what advice the AI is actually giving?
There’s also the issue of content quality. AI tutors can make mistakes, give generic or even misleading advice, or reinforce bad habits if not carefully monitored. In the study, some kids found the AI’s suggestions helpful, while others got confused or frustrated by unclear feedback. For families already worried about too much screen time, the idea of adding another digital voice to the mix may not be reassuring.
Teachers, too, face new challenges. AI tutors could help lighten the load in crowded classrooms, but they also risk replacing human interaction with automated responses. And if something goes wrong—say, the AI gives out-of-date or inappropriate advice—who’s responsible?
The Bigger Picture
The move toward voice- and screen-sharing AI tutors is part of a larger trend in education, where tech companies are racing to automate learning. In 2025, the U.K. Department for Education piloted AI-powered math tutors in 200 schools, reporting mixed results: some students improved, but others struggled with the AI’s one-size-fits-all approach. Meanwhile, U.S. lawmakers have yet to update COPPA or pass new rules specifically for AI in classrooms, despite calls from groups like the American Federation of Teachers for stricter oversight.
Big tech’s push into schools isn’t just about learning—it’s also about data. AI tutors collect massive amounts of information about how kids think, what mistakes they make, and even how they talk. In 2024, a report by the Center for Democracy & Technology warned that "AI-powered edtech tools risk exposing students to unprecedented surveillance." Without clear rules, parents and schools are left to figure out what’s safe on their own.
Some experts argue that AI tutors could help close learning gaps, especially for students who don’t get enough one-on-one support. But others worry about over-reliance on technology, especially when it comes to young children still developing critical thinking and social skills. For more on how to prepare kids for a future shaped by AI, see our guide on what skills kids need for the future.
What You Can Do
- Ask questions before signing up: If your child’s school or after-school program introduces an AI tutor, ask how it works, what data it collects, and who can see that information. Don’t be afraid to request a demonstration or written policy.
- Monitor AI interactions: Spend time watching how your child uses the AI tutor. Listen in on a session, or review transcripts if available. Make sure the advice is accurate and age-appropriate.
- Set boundaries for screen time: AI tutors can be helpful, but they shouldn’t replace human interaction or hands-on learning. See our tips on managing screen time for kids.
- Teach your child to question AI: Encourage kids to double-check the AI’s advice, ask for help from a real person when stuck, and never share personal information with the AI.
- Stay informed about AI safety: Follow trusted sources and parent guides on keeping kids safe from AI. Laws and best practices are changing fast.
The Bottom Line
AI tutors that talk and watch through screens are already changing how kids learn, for better and for worse. While they can offer real help, they also raise tough questions about privacy, quality, and the role of adults in education. As these tools become more common, parents and teachers will need to stay vigilant, ask hard questions, and push for stronger safeguards—so that technology helps kids, not the other way around.
New AI Safety Framework Aims to Protect Kids in Schools
A new safety framework for educational AI could help shield students from harmful content, but experts warn it's early days. Parents and teachers should stay cautious.
Read analysis ENVIRONMENT ENERGYSt. Joe County Residents Fear Data Center Will Worsen Flooding
St. Joe County residents worry a new AI data center’s water pumping will worsen local flooding, raising safety and property concerns for families.
Read analysis PARENTING EDUCATIONAI Tools Undermine College Testing, Forcing Big Changes
AI-powered cheating is shaking up college assessments, pushing schools to overhaul how they test students. Families should prepare for big changes.
Read analysis