TL;DR
Get business pricing on monitors, keyboards and dev gear
- Business-only prices and quantity discounts
- Tax-exempt purchasing
- Multiple users, one account, clear invoices
Google has published findings from two studies of Gemini in education: a 10-week UK math-tutoring trial involving more than 1,100 students and a yearlong study across 31 Italian schools. The company reports similar performance for AI- and human-tutored students on classroom tasks and quizzes, and more collaborative lessons in Italy; the UK trial found no gains on STAR standardized-test scores.
Google on October 29, 2026, reported results from two studies examining Gemini in schools: a 10-week math-tutoring trial with more than 1,100 students in the UK and a yearlong study across 31 schools in Italy. The company said UK students using AI tutoring performed similarly to students working with human tutors on classroom tasks and quizzes, while Italian teachers reported more active, collaborative lessons when using Gemini to plan. The UK trial did not find gains on STAR standardized-test scores.
The UK randomized controlled trial involved students aged 12 to 15, who used Eedi’s school platform. When students answered questions incorrectly, they received either a static hint, help from a human tutor or interactive AI tutoring that combined Eedi’s teaching resources with Gemini. Google said teachers monitored students and provided classroom support, while tutors also guided students through the platform. Human experts reviewed every AI response in real time before it reached a student.
Google reported that expert reviewers approved 95% of AI responses, up from 76% in a prior study. Students receiving AI tutoring performed similarly to those with human tutors on platform tasks and intermediate quizzes. The company said Gemini guided students rather than supplying answers in nearly 99% of sessions, and students worked through the math themselves in more than 85% of chats. These are findings reported by Google, not independent evaluations presented in the announcement.
The Italian study was conducted with the Salesians of Don Bosco school network and the Pontifical Salesian University in Rome. It included 1,600 teachers and represented more than 24,500 students aged 13 to 18. Google said teachers who used Gemini to co-design lessons reported greater creativity than those planning without it, and that traditional lectures declined while group work and peer discussion increased. The announcement describes teacher reports and classroom activity; it does not establish that Gemini alone caused those changes.
Classroom Results and Their Limits
The findings speak to two practical questions for schools: whether AI can provide useful individual support when a teacher cannot sit with every student, and whether it can help teachers prepare lessons that involve more student participation. Google’s UK results suggest Gemini-supported tutoring may match human tutoring on some classroom measures under the trial’s conditions, including live expert review and teacher involvement. The Italian results point to a possible role in lesson planning, rather than replacing educators in instruction.
But the distinction between classroom performance and broader learning outcomes matters. The tutoring group completed fewer problems per session than students receiving static hints, and Google reported no improvement on STAR standardized tests during the 10-week trial. Schools weighing adoption need evidence not only about engagement or quiz performance, but also about learning over longer periods, the resources required for oversight and whether results hold across different settings.
Popular right now
Bose QuietComfort Wireless Bluetooth Headphones with Active Noise Cancellation, Mic, USB‑C Charging, Deep Bass, 24‑Hour Battery – White Smok
Bose QuietComfort True Wireless Earbuds with Active Noise Cancellation, Black
Apple AirPods Pro 3 Wireless Earbuds with Active Noise Cancellation
As an affiliate, we earn on qualifying purchases.
How the Two Studies Were Run
Google said the research followed earlier exploratory work with education platform Eedi, which examined AI tutoring in real classrooms. The new UK trial compared three kinds of help after a wrong answer: a static hint, a human tutor or interactive AI tutoring. Because experts checked each AI response before students saw it, the reported results reflect a supervised setup—not an unsupervised chatbot used independently by students.
In Italy, the research extended an earlier pilot in northern Italy into a yearlong study involving primary, secondary and vocational educators. The company said researchers collected classroom activities and teacher self-reports about lesson planning with and without Gemini for Education. Google’s announcement reports that lecture-style teaching fell by 38% and student-led collaborative learning nearly tripled in lessons teachers co-designed with Gemini. Those figures describe the study’s reported classroom patterns; they do not, on their own, measure long-term academic attainment.
“AI tutoring using Gemini capabilities can deliver support on par with human tutors for UK students when used responsibly with human oversight.”
— Google, in its report
Long-Term Learning Still Unproven
The UK trial lasted 10 weeks, and Google said the AI-tutored students solved about 14 problems per session, compared with 19 for students using static hints. The company suggested that the slower, step-by-step approach meant students covered less curriculum, which may help explain why classroom-task gains did not appear in STAR scores. That explanation is Google’s interpretation; the reported trial does not establish that it was the sole reason for the standardized-test result.
The announcement does not provide enough detail to assess all study methods, effect sizes or the full range of results independently. It also does not establish whether the Italian classroom changes led to measurable improvements in student achievement or persisted beyond the study. The findings should be read as company-reported evidence from particular school settings, not as proof that similar outcomes will occur everywhere.
Yearlong Tutoring Trials Planned
Google said it is exploring follow-up tutoring trials that would run across a full academic year and use expanded measures to examine what drives learning and later outcomes. A longer study could test whether students cover enough curriculum while receiving step-by-step help, and whether gains on assignments and quizzes eventually translate into standardized-test results.
The company has not specified when those trials will begin or when results will be published. For now, the reported evidence centers on supervised tutoring in the UK and teacher-reported classroom planning in Italy. Further independent research and clearer information about study methods, costs and outcomes will be needed to judge how widely schools can apply the findings.
Key Questions
What did the UK Gemini tutoring trial find?
Google said students using Gemini-supported tutoring performed similarly to students paired with human tutors on platform assignments and intermediate quizzes. The trial did not find gains on STAR standardized-test scores over its 10-week period.
Did Gemini replace teachers or tutors in the UK study?
No. Teachers monitored students and offered classroom support, tutors guided students through the platform, and human experts reviewed every AI response before it reached a student.
What did the Italian study report?
Google said teachers using Gemini to co-design lessons reported greater creativity, while lecture-style teaching fell and collaborative activities increased. The findings came from a yearlong study across 31 schools and included teacher self-reports and classroom activity.
Does the research prove Gemini improves student achievement?
No. Google reported results on classroom tasks and quizzes in the UK, but no STAR test gains during the trial. The Italian report describes changes in teaching and classroom activity, not proof of improved long-term academic achievement.
What research is planned next?
Google said it is exploring tutoring trials lasting a full academic year and plans to broaden the measures used to study learning outcomes. It has not announced a schedule for those trials or publication of results.
Source: blog.google
