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 a 10-week UK trial of Gemini-supported math tutoring and a yearlong study in Italian schools. The company reports that students receiving AI tutoring performed similarly to those paired with human tutors on platform tasks and quizzes, while standardized-test gains were not detected; Italian educators reported more collaborative lessons when co-designing with Gemini.
Google published findings on October 29 from studies of Gemini in UK and Italian classrooms, reporting similar performance between AI- and human-tutored students on some UK math measures and more collaborative teaching practices in Italy. The results come with limits: the UK trial found no gains on STAR standardized tests during its 10-week period, and the Italian findings are based on collected classroom activities and teacher self-reports.
The UK randomized controlled trial involved more than 1,100 students aged 12 to 15 using the Eedi school platform. Over 10 weeks, students who answered questions incorrectly received either a static hint, help from a human tutor, or interactive AI tutoring that combined Eedi’s resources with Gemini capabilities. Teachers monitored progress and supported students, while tutors also provided one-to-one guidance through the platform.
Google says students receiving AI tutoring performed similarly to students paired with human tutors on Eedi assignments and intermediate quizzes. Expert reviewers approved 95% of AI responses outright, up from 76% in the previous study, a nearly 25% improvement in pedagogical quality as reported by Google. The company also says the tutor guided students rather than giving answers in nearly 99% of sessions; students worked through the math themselves in more than 85% of chats. Experts reviewed every AI response in real time before students saw it.
In Italy, a yearlong study covered 31 schools, 1,600 teachers and more than 24,500 students aged 13 to 18 in the Salesians of Don Bosco school network. Google reports that teachers who used Gemini to co-design lessons said they felt more creative in planning and teaching than when they did not use it: 79% compared with 30%. The company also reports that traditional front-of-class lectures fell by 38%, while student-led collaborative learning nearly tripled. These are reported study findings, not evidence that Gemini alone caused the changes.
Tutoring Results Come With Trade-Offs
The findings point to two different classroom uses for generative AI: individual academic support and help for teachers designing more active lessons. For schools considering tutoring tools, the UK results suggest an AI system can support practice and quiz performance at a level comparable to human tutoring under the study’s conditions, including live expert review and teacher involvement.
But the lack of a measured STAR test gain matters. Google says students solved about 14 problems per session with AI tutoring, compared with 19 using static hints, because step-by-step help took longer. The company’s explanation is that students covered less curriculum in the trial, limiting what could be measured on a broad standardized test. That is an interpretation offered by Google; the trial does not establish whether longer use would produce standardized-test gains.
For teachers, reported changes in Italian classrooms may indicate that AI can be used as a planning aid rather than as a replacement for educators. The study’s scope and self-reported measures mean the results should be read as evidence from participating schools, not a guarantee of the same effects elsewhere.
Products Worth Considering
EcoFlow Smart Home Panel 2 Whole-Home Backup System Hub
Logitech MX Brio Ultra HD 4K Webcam, Graphite
Sony WH-1000XM6 Wireless Noise-Canceling Over-Ear Headphones, Blue, Bundle
As an affiliate, we earn on qualifying purchases.
How the Classroom Studies Were Run
Google says the UK trial followed an earlier exploratory study with Eedi and was designed to examine student learning over time in structured math lessons. It compared three forms of help after an incorrect answer and tracked student progress through platform assignments, intermediate quizzes and independent STAR standardized tests. Teachers remained involved, and experts checked AI responses before delivery.
The Italian work followed an initial pilot in Northern Italy and was conducted with the Salesians of Don Bosco school network and the Pontifical Salesian University in Rome. Across the year, researchers collected classroom activities and teacher self-reports about lesson planning with and without Gemini for Education. The different study designs and measures mean the UK tutoring results and Italy classroom reports cannot be treated as a single test of Gemini’s overall educational impact.
“Technology best serves education when it creates space for inspiring teaching and deeper learning.”
— Google, in its October 29, 2026 report
Questions Beyond the Trial Period
The UK study did not show gains on STAR standardized tests over the 10-week trial, despite reported improvements on daily tasks and intermediate quizzes. Google attributes the difference to slower, step-by-step tutoring and lower problem volume, but says longer trials are needed to examine that explanation. The published summary does not establish whether the same pattern would hold across a full school year or in other subjects and age groups.
The Italian results also leave questions about how broadly the findings apply. Google describes teacher self-reports and collected classroom activities, but the report summary does not provide enough detail to determine how classroom changes were independently measured or how participating schools compare with other schools. Independent replication and fuller study methods would help clarify the size and durability of the reported effects.
Longer Trials and Further Measures
Google says it is exploring follow-up UK trials across a full academic year and expanding the metrics used to study learning and longer-term outcomes. Those studies could test whether students cover enough curriculum with step-by-step AI tutoring to affect broader standardized assessments, while examining how results vary across learners and classrooms.
The company has also pointed readers to a full technical report for more detail on the UK trial. For schools weighing similar tools, the next evidence to watch includes study methods, independent evaluation, student outcomes over longer periods, and whether the safeguards used in this trial—especially real-time expert review—are practical in routine classroom use.
Key Questions
What did the UK Gemini tutoring trial find?
Google says students using Gemini-supported tutoring performed similarly to human-tutored students on platform tasks and intermediate quizzes. The trial did not find gains on STAR standardized-test scores over 10 weeks.
Did AI tutoring improve standardized-test scores?
Not in the reported UK trial period. Google says gains on classroom assignments and quizzes did not appear on STAR tests during the 10 weeks; longer-term effects remain unknown.
How was student safety and tutoring quality handled?
According to Google, human experts reviewed every AI response in real time before it reached a student. Teachers also monitored progress and provided classroom support.
What did the study in Italy report?
Across 31 schools, teachers reported greater creativity when co-designing lessons with Gemini. Google also reported fewer traditional lectures and nearly three times as much student-led collaborative learning, based on the study’s collected classroom activities and teacher self-reports.
What happens next?
Google says it is exploring full-academic-year follow-up trials in the UK and broader measures of learning outcomes. The company has not reported those results yet.
Source: blog.google
