Artificial intelligence (AI) is fast becoming a regular part of education practice, offering new ways to adapt instruction, assess student learning in real time, tailor feedback, and support decision making. For education leaders working to improve underperforming schools, understanding how AI is being used—and the evidence behind it—may help unlock strategies to improve teaching quality, personalize learning, and accelerate student achievement.
Through its Ask an Expert service, REL Central responds to requests from state and local education leaders for quick-turnaround, evidence-based information on topics that address priority needs in the Central region. Our recent Ask an Expert addressed a state partner’s request for information about potential uses of AI for school turnaround. The Ask an Expert highlights emerging AI applications and summarizes what current research tells us about their effectiveness.
AI in Action: How States and Districts Are Leading the Way
Although their approaches and applications vary, states and districts across the country are experimenting with AI-driven tools to enhance teaching, personalize learning, and inform data-driven decisions—all of which have the potential to support school turnaround. At the state level, agencies in Massachusetts and Connecticut are piloting AI-enabled curricula and tools to support teachers as they become familiar with AI.[1],[2], [3] Similarly, Texas, Kentucky, and New Mexico are deploying predictive analytic tools to forecast student performance and reveal early signs of risk; this can generate faster insights and guide intervention planning.[4], [5], [6] In Iowa, Arizona, and Indiana, school districts are piloting AI-based tutoring tools to deliver personalized academic support.[7], [8], [9], [10]
School districts and states have begun sharing some initial findings on the effectiveness of AI tools in addressing chronic absenteeism, supporting teacher instruction, and enhancing student learning. For example, in New Mexico, a text-messaging AI tool may have increased parent-school communication and slightly increased attendance rates, but outdated student data and inconsistent parent engagement were challenges to implementation.[11] In Indiana, just over half of surveyed teachers reported that an AI tutoring platform had a positive or very positive impact on their teaching practice and on student learning, whereas 40 percent of teachers reported no changes.[12]
Additional state and district AI initiatives or pilots are taking place during the 2025/26 school year and related research efforts are underway. As states and districts continue to implement and evaluate new AI tools, more research on their effectiveness is on the horizon.
Insights from Emerging Research
Although AI applications are evolving rapidly, emerging findings can generate helpful insights. In one rural context, for example, deep learning models helped optimize how schools allocated resources, staff, and material.[13],[14] In other studies, machine-learning early warning systems yielded mixed results—they predicted near-term academic risks but may not be more efficient in doing so than traditional models are.[15],[16] Additional research has shown that educators can benefit from AI-supported coaching and lesson-planning tools, which show promise in improving instructional quality and reducing workload.[17],[18]
The strongest evidence comes from student-facing AI tools. Systematic reviews covering more than 80 causal studies found positive effects on student engagement, learning behaviors, and cognitive outcomes.[19], [20], [21] Math-focused AI tutors helped students answer more questions correctly and spend more time practicing skills,[22],[23] whereas tools in non-math subjects showed improvements in grammar, motivation, and learning processes.[24], [25], [26] Moreover, students working with AI tutors and chatbots were more engaged, and this fostered deeper learning.[27],[28],[29] Although assessment gains varied, improvements in engagement and learning processes were consistent.
What This Means for School Turnaround
The evidence suggests that AI can support school turnaround by providing timely, actionable data; expanding access to personalized tutoring; and enhancing teachers’ professional learning. Predictive models may help identify schools and students needing additional support, and AI tutors can extend instructional capacity without large staffing increases. Teacher-support tools may strengthen instruction in cost-effective and scalable ways.
At the same time, studies highlight that AI is most effective when paired with strong human judgment and high-quality instruction. To effectively use AI, educators and education leaders need the resources and professional learning to improve digital literacy and support a deeper understanding of AI in education. Moreover, students need a better understanding of what AI is and how it should and could be used. AI should be seen as a complement to—not a replacement for—the essential work of educators.
For states exploring AI for school turnaround, next steps may include identifying priority use cases, piloting tools aligned with their strategic priorities, and assessing the infrastructure and training needed for responsible implementation. As the use of AI in education grows, states have an opportunity to integrate these tools in ways that enhance instruction, deepen student engagement, and support meaningful and sustainable improvement.
[1] Healey-Driscoll Administration. (2024, September 18). Future Ready: AI in the Classroom for Educators initiative launched to support teachers statewide. Commonwealth of Massachusetts. https://www.mass.gov/news/healey-driscoll-administration-launches-future-ready-ai-in-the-classroom-for-educators
[2] Massachusetts Executive Office of Education. (2024, March 19). New AI curriculum pilot to reach 1,600 Massachusetts students across 30 school districts. https://www.mass.gov/news/new-ai-curriculum-pilot-to-reach-1600-massachusetts-students-across-30-school-districts
[3] Connecticut State Department of Education. (2024). Connecticut AI in Education Pilot Program. GoOpenCT. https://goopenct.org/groups/connecticut-ai-in-education-pilot-program/102/
[4] Texas A&M University Department of Educational Psychology. (n.d.). District School Performance Predictor. https://dspp.education.tamu.edu/about
[5] Kentucky Department of Education. (n.d.). Early warning and persistence to graduation. https://www.education.ky.gov/educational/int/Pages/EarlyWarningAndPersistenceToGraduation.aspx
[6] New Mexico Public Education Department. (2024, October 4). New Mexico schools use AI to track student absences. GovTech. https://www.govtech.com/education/k-12/new-mexico-schools-use-ai-to-track-student-absences
[7] Iowa Department of Education. (2024, August 27). Iowa Department of Education launches new personalized reading tutor in Iowa schools. https://educate.iowa.gov/press-release/2024-08-27/iowa-department-education-launches-new-personalized-reading-tutor-iowa-schools-builds-upon-prior
[8] Chaviano, S. P. (2025). Finding their voice: How Amira is helping Aldine students read with confidence. Aldine Independent School District. https://www.aldineisd.org/2025/09/17/finding-their-voice-how-amira-is-helping-aldine-students-read-with-confidence/
[9] Arizona Department of Education. (2024). Arizona Department of Education Learning (ADEL) portal. https://adel.azed.gov/home
[10] Indiana Department of Education. (2024). AI-powered platform pilot grant final report. https://drive.google.com/file/d/13eJJ5dICQqvD7q2l2qWwhp4t8vuec_nt/view
[11] New Mexico Public Education Department. (2024, October 4). New Mexico schools use AI to track student absences. GovTech. https://www.govtech.com/education/k-12/new-mexico-schools-use-ai-to-track-student-absences
[12] Indiana Department of Education. (2024). AI-powered platform pilot grant final report. https://drive.google.com/file/d/13eJJ5dICQqvD7q2l2qWwhp4t8vuec_nt/view
[13]Zhao, X. (2025). Using deep learning to optimize the allocation of rural education resources under the background of rural revitalization. Journal of Combinatorial Mathematics and Combinatorial Computing. https://doi.org/10.4018/IJAEIS.375426
[14] Liu, Y. (2025). Research on deep learning algorithm application and resource allocation optimization in educational resources big data analysis. Journal of Combinatorial Mathematics and Combinatorial Computing, 127b, 6539–6555. https://doi.org/10.61091/jcmcc127b-358
[15] Cattell, L., & Bruch, J. (2021). Identifying students at risk using prior performance versus a machine learning algorithm (REL 2021–126). U.S. Department of Education, Institute of Education Sciences, National Center for Education Evaluation and Regional Assistance, Regional Educational Laboratory Mid-Atlantic.
[16] Bruch, J., Gellar, J., Cattell, L., Hotchkiss, J., & Killewald, P. (2020). Using data from schools and child welfare agencies to predict near-term academic risks (REL 2020–027). U.S. Department of Education, Institute of Education Sciences, National Center for Education Evaluation and Regional Assistance, Regional Educational Laboratory Mid-Atlantic. https://files.eric.ed.gov/fulltext/ED606230.pdf
[17] Yousef, J., Ibrahim, M. K., Alsalhi, N. R., Alqawasmi, A. A., & Ahmed, A. (2024). Utilizing AI-driven virtual training platforms to enhance smart board skills among English language teachers. Eurasian Journal of Applied Linguistics, 10(3), 175–184. https://files.eric.ed.gov/fulltext/EJ1475073.pdf
[18] Copur-Gencturk, Y., Li, J., & Atabas, S. (2024). Improving teaching at scale: Can AI be incorporated into professional development to create interactive, personalized learning for teachers? American Educational Research Journal, 61(4), 767–802. https://doi.org/10.3102/00028312241248514
[19] Younas, M., Ismayil, I., El-Dakhs, D. A. S., & Anwar, B. (2025). Exploring the impact of artificial intelligence in advancing smart learning in education: A meta-analysis with statistical evidence. Open Praxis, 17(3), 594–610. https://doi.org/10.55982/openpraxis.17.3.842
[20] García-Martínez, I., Fernández-Batanero, J. M., Fernández-Cerero, J., & León, S. P. (2023). Analysing the impact of artificial intelligence and computational sciences on student performance: Systematic review and meta-analysis. Journal of New Approaches in Educational Research, 12(1), 171–197. https://doi.org/10.7821/naer.2023.1.1240
[21] Gunsaldi, M. S., Guner, E. G., Uckan, M., & Bati, K. (2025). The impact of generative AI applications on student learning outcomes in science education: A systematic review. Journal of Education in Science, Environment and Health, 11(3), 196–208. https://doi.org/10.55549/jeseh.840
[22] Henkel, O., Horne-Robinson, H., Kozhakhmetova, N., & Lee, A. (2025). Effective and scalable math support: Experimental evidence on the impact of an AI math tutor in Ghana. Proceedings of the 19th International Conference on Artificial Intelligence in Education (AIED 2025). University of Oxford and J-PAL North America.
[23] Liu, J., Sun, D., Sun, J., Wang, J., & Yu, P. L. H. (2025). Designing a generative AI enabled learning environment for mathematics word problem solving in primary schools: Learning performance, attitudes, and interaction. Computers and Education: Artificial Intelligence, 9, Article 100438. https://doi.org/10.1016/j.caeai.2025.100438
[24] Vanzo, A., Pal Chowdhury, S., & Sachan, M. (2024, September 24). GPT-4 as a homework tutor can improve student engagement and learning outcomes. arXiv. https://doi.org/10.48550/arXiv.2409.15981
[25] Eteng-Uket, S., & Ezeoguine, E. (2025). The impact of artificial intelligence chatbots on student learning: A quasi-experimental analysis of learning outcome and engagement. Journal of Educators Online, 22(2), 1–15. https://files.eric.ed.gov/fulltext/EJ1470567.pdf
[26] Kara, S. (2025). The effect of artificial intelligence applications in 6th grade visual arts course on student attitudes and course outcomes. International Journal of Modern Education Studies, 9(1), 51–82. https://doi.org/10.51383/ijonmes.2024.405
[27] Younas, M., Ismayil, I., El-Dakhs, D. A. S., & Anwar, B. (2025). Exploring the impact of artificial intelligence in advancing smart learning in education: A meta-analysis with statistical evidence. Open Praxis, 17(3), 594–610. https://doi.org/10.55982/openpraxis.17.3.842
[28] García-Martínez, I., Fernández-Batanero, J. M., Fernández-Cerero, J., & León, S. P. (2023). Analysing the impact of artificial intelligence and computational sciences on student performance: Systematic review and meta-analysis. Journal of New Approaches in Educational Research, 12(1), 171–197. https://doi.org/10.7821/naer.2023.1.1240
[29] Gunsaldi, M. S., Guner, E. G., Uckan, M., & Bati, K. (2025). The impact of generative AI applications on student learning outcomes in science education: A systematic review. Journal of Education in Science, Environment and Health, 11(3), 196–208. https://doi.org/10.55549/jeseh.840
