TL;DR
Get the latest gadgets delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
Google says studies in the UK and Italy found Gemini-supported tutoring performed similarly to human tutoring on platform tasks and quizzes, while teachers using Gemini to plan lessons reported more creative planning and more collaborative classroom activity. The UK trial did not find improved scores on STAR standardized tests during its 10-week period, and Google says longer-term research is needed.
Google published new findings on October 29 from studies of Gemini in schools in the UK and Italy, reporting that AI tutoring performed similarly to human tutoring on some classroom measures and that teachers using Gemini to plan lessons reported more active, collaborative teaching. The results come from a 10-week UK trial and a yearlong study across 31 Italian schools; Google says the UK trial did not show a gain on STAR standardized test scores during the study period.
In the UK, Google and education platform Eedi ran a randomized controlled trial involving more than 1,100 students aged 12 to 15. Students received a static hint, help from a human tutor or interactive AI tutoring that combined Gemini capabilities with Eedi’s teaching resources. Teachers monitored progress and supported students in class. According to Google, experts reviewed every AI response in real time before it reached a student.
Google reports that students using AI tutoring performed similarly to students working with human tutors on Eedi platform tasks and intermediate quizzes. Expert reviewers’ outright approval of AI responses rose from 76% in a previous study to 95% in this one, which Google describes as a nearly 25% improvement in pedagogical quality. The company also says Gemini guided students rather than directly giving answers in nearly 99% of sessions, while students worked through the mathematics themselves in more than 85% of chats.
The Italian study covered 1,600 teachers and more than 24,500 students aged 13 to 18 in primary, secondary and vocational education across a network of 31 schools. Google says teachers reported feeling more creative in lesson planning with Gemini than without it, at 79% compared with 30%. The company also reports that conventional front-of-class lectures fell by 38%, while workshops, group activities and peer discussion increased. These are findings reported by Google; the supplied account does not provide further methodological detail or independent evaluation.
Classroom Gains and Testing Limits
The findings speak to two practical questions for schools: whether AI can provide useful academic help at scale, and whether it can give teachers more room to organize active learning. Google’s UK results suggest AI tutoring may support performance on routine assignments and quizzes when paired with teacher involvement and review of responses. The Italian results point to a possible change in classroom practice when educators use Gemini as a planning partner rather than a replacement for instruction.
The limits matter as much as the reported gains. The UK study found no improvement on STAR standardized test scores over 10 weeks. Google says students using step-by-step AI tutoring completed about 14 problems per session, compared with 19 for students receiving static hints. The company attributes the test-score result in part to students covering less curriculum material during the trial. That explanation is Google’s interpretation, not proof that the shorter problem volume caused the result.
For schools weighing AI tools, the research offers a case for supervised use, but not evidence that Gemini improves every kind of learning outcome. Results from classroom tasks, quizzes and standardized tests are distinct measures, and the reported studies do not establish that the outcomes will carry over to other schools or age groups.
Products Worth Considering
Bose TV Speaker (2nd Gen) Compact Soundbar with Bluetooth and HDMI
Anker SOLIX S2000 Portable Power Station, 2,010Wh, 1,500W
Apple iPad Air 11-inch (M2), 512GB, Wi-Fi, Space Gray
As an affiliate, we earn on qualifying purchases.
How the Two School Studies Worked
The UK research follows an earlier exploratory study with Eedi. In the newer trial, students who answered a question incorrectly received one of three forms of support: a fixed hint, assistance from a human tutor or interactive AI tutoring. Google says the trial tracked progress through platform assignments, intermediate quizzes and independent STAR tests. Teachers remained involved, and human experts checked every AI response before delivery.
Google’s comparison with the previous study concerns reviewers’ approval of AI responses, rising from 76% to 95%. This is a reported change in approval rate, not a measure of student test-score improvement. The Italian research followed an initial pilot in Northern Italy and expanded to a yearlong study with the Salesians of Don Bosco school network and the Pontifical Salesian University in Rome. Teachers reported on lesson planning and classroom activity with and without Gemini for Education.
Both studies were conducted with Google’s involvement, and the findings were published by the company’s researchers. The UK trial included a comparison group and randomized design, while the Italian account describes classroom activities and teacher self-reports. Those differences affect what each study can show: the UK trial measured student outcomes under specified conditions, while the Italian findings report changes and experiences associated with Gemini-supported lesson planning.
“Technology best serves education when it creates space for inspiring teaching and deeper learning.”
— Google’s study authors
What the Findings Cannot Establish
The report does not establish whether the UK tutoring results would persist over a full school year or produce stronger standardized-test outcomes. Google says the 10-week period may have been too short for students to cover the breadth of material on the STAR test, and it is examining that possibility in follow-up research. Whether that explanation accounts for the lack of score gains remains uncertain.
The supplied report also does not give detailed results for every measure in the Italian study, including the extent of student response beyond the reported increase in collaborative activity. Teacher self-reports and recorded classroom activities can describe classroom experience, but the account does not establish that Gemini alone caused the reported changes. The findings should not be read as proof that the same results will occur across different schools, subjects or teaching approaches.
Google Plans Longer-Term Research
Google says it is exploring follow-up UK trials that would run across a full academic year. The company also plans to broaden the measures it uses to understand what supports meaningful learning and eventual outcomes. Further results could help test whether students who complete fewer problems per session with step-by-step tutoring later catch up on curriculum coverage or show changes on standardized assessments.
The October report does not specify a timetable for those trials or announce a next publication date. Until additional findings are available, the current evidence is limited to the study periods and school settings described by Google. For educators, the immediate takeaway is that the reported benefits were observed alongside human oversight, teacher involvement and differences between classroom measures and standardized-test results.
Key Questions
What did the UK Gemini tutoring trial find?
Google reports that students using Gemini-supported tutoring performed similarly to students receiving human tutoring on Eedi assignments and intermediate quizzes. The trial did not find improved STAR standardized-test scores over its 10-week period.
Did Gemini replace teachers or human tutors in the UK study?
No. Teachers monitored student progress and supported classes, and human experts reviewed every AI response before students received it. The trial compared AI tutoring with human tutoring and static hints as different forms of support.
What did teachers in Italy report?
Google says 79% of teachers using Gemini reported feeling more creative in teaching and planning, compared with 30% without Gemini. The company also reported fewer front-of-class lectures and more collaborative activities in lessons planned with Gemini.
Why were STAR test scores unchanged in the UK trial?
Google suggests step-by-step tutoring took longer per question, so students completed about 14 problems per session compared with 19 for students receiving static hints. The company says the lower volume may have limited curriculum coverage, but the report does not establish that this caused the unchanged scores.
When will more results be available?
Google says it is exploring UK follow-up trials lasting a full academic year, but the report does not give a schedule or publication date for further findings.
Source: blog.google
