The "AI Dividend"—Does AI Create More Time for Teachers?

An early analysis of Gallup's newly released study.

The AI “Dividend”

Gallup (with support from the Walton Family Foundation) is out today with a new study, “Teaching for Tomorrow: Unlocking Six Weeks a Year With AI”, on how teachers use AI. The results are based on a nationally representative poll (using RAND’s American Educator panel as the sample) and interviews with a subset of respondents.

Three in ten teachers in the study say they use AI weekly, saving, on average, six hours a week —or six weeks a year—of time by using AI tools for lesson planning, grading, and worksheets. Teachers say they reinvest that time in varied ways. Sixty percent say they have more time for improving teaching and learning, and more time to give individualized feedback and support to students.

Gallup: “Three in 10 Teachers Use AI Weekly, Saving Six Weeks a Year”

Beyond the big headline, the Gallup findings confirm much of the research we’ve conducted at CRPE, including our study of AI Early Adopter school districts and results from our American School District Panel teacher survey with RAND.

• Six in 10 teachers say they have used AI for their work this year.

• Teachers are generally more supportive of than opposed to AI in education. Overall, 40% of teachers strongly or somewhat favor using AI tools in K-12 schools, and 28% strongly or somewhat oppose.

• Only 1 in 3 teachers report getting training in AI from their school district.

• Half of the teachers polled do not give their students much direction at all about AI.

• Only 19% say their school has an AI policy. Teachers in schools with a policy are more likely to use AI, and their schools received a 26% greater “AI dividend,” or more hours saved per teacher.

• Teachers who use AI are also more likely to have a positive view about the impacts of AI on student outcomes: 48% of weekly AI users think AI will increase student engagement, compared with 25% of non-users.

• Teachers are most optimistic about the potential for AI to support accessibility: 57% of teachers agree (46%) or strongly agree (11%) that AI will improve the accessibility of learning materials for students with disabilities. Special education teachers are even more likely to agree that AI will yield this benefit (65%).

• Most teachers who frequently use AI tools say the tools improve the quality of their work: 64% see higher quality in the modifications they make to student materials, 61% say they generate higher-quality insights about student learning or achievement data, and 57% say AI improves the quality of their grading and student feedback.

And for those of us who have been asking ed tech developers to be more responsive to the needs of educators: teachers have a strong interest in using AI for a variety of purposes, but there is a gap between what they say they want to use AI for (better use of data and assessments, adapting curriculum to students needs, etc.) and what they are currently using it for. In CRPE’s research (we have new report on this coming soon), teachers said they can get overwhelmed by the complexity and number of AI tools available and frustrated that the tools do not speak to each other. More work is needed to understand the gap between user demand and supply.

Chart used with permission from Gallup.

The study results are an essential first deep dive into AI and teacher productivity and use. But these are self-reports, and the interview data are not necessarily representative. Nor are reported relationships—say, between policy and practice—necessarily causal. We need more research on AI and teacher productivity, but these numbers are a really interesting start at understanding how U.S. teachers are using AI.

Final Words

“Schools that are unable to support their teachers with training and school-wide policies on AI may risk widening the gaps that already exist between teachers with and without AI skills.”

-Gallup’s Teaching for Tomorrow report