CoChat
Sign Up
Cited by 30Student Assessment and FeedbackArtificial Intelligence in EducationIntelligent Tutoring Systems and Adaptive Learning

“We Should Not Be Like a Dinosaur”—Using AI Technologies to Provide Formative Feedback to Students

Open access
Burner, Tony, Lindvig, Yngve, Wærness, Jarl Inge·2025·Education Sciences
DOI 10.3390/educsci15010058
Open this paper in CoChat — free accountOpen PDF

Abstract

Artificial intelligence (AI) technologies have the potential to enhance learning, teaching, and assessment by providing AI-generated feedback to students. This study used five different AI plug-ins and four different knowledge bases for the optimization of feedback in classroom formative assessments. The sample consisted of 26 students and 13 teachers at five secondary schools in Norway. Interviews and unstructured observations were used to collect data. The findings indicate that AI-generated feedback is appreciated by both students and teachers. It provides immediate subject-specific, concrete, and relevant feedback. High-performing students were motivated to further improve their writing. However, some students found the feedback to be too general or complicated, and the teachers who had not conducted whole-term periodic planning struggled with using AI for formative assessment purposes. Finally, both students and teachers contend that the teacher needs to have the last word when AI takes over formative feedback procedures.

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.3390/educsci15010058

Cites

19 of 19 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • The feedback literacy and self-regulated learning objective aims to strengthen students' capacity to understand, act on, and self-regulate in response to AI-generated feedback.

    “This objective aims to (1) strengthen students' capacity to understand, act on, and selfregulate in response to AI-generated feedback [124-127];”

    From A Systematic Review of Generative AI in K–12: Mapping Goals, Activities, Roles, and Outcomes via the 3P Model (2025) · cites this paper as [124]

    The Theoretical Framework (page not provided) says formative feedback tells students how to improve and that students need to act on it. Section 2.1.3 (page not provided) describes AI feedback supporting continuous learning and reflection. The paper does not identify a feedback-literacy or self-regulated-learning objective, or establish that its AI feedback strengthens students’ capacity to self-regulate. (LLM, conf=0.82)

    Experimental — accuracy not yet measured

  • AI adjusts the specificity, depth, and difficulty of feedback based on students' abilities and prior performance, enabling tailored prompts and tiered guidance.

    “AI adjusts the specificity, depth, and difficulty of feedback based on students' abilities and prior performance, enabling tailored prompts and tiered guidance.”

    From A Systematic Review of Generative AI in K–12: Mapping Goals, Activities, Roles, and Outcomes via the 3P Model (2025) · cites this paper as [124]

    Section 2.1.3 says the AI offers personalized feedback based on a student’s current progress and suggests new approaches or deeper exploration. The paper does not establish that it systematically adjusts feedback specificity, depth, and difficulty according to ability and prior performance, or that it provides tiered guidance; Section 3.2.3 reports that some feedback was too unspecific, hard, or long. Page numbers were not available in our copy. (LLM, conf=0.84)

    Experimental — accuracy not yet measured

  • A large subset of studies casts GenAI as a grader in writing and reading assessments, delivering analytic diagnoses and scores across dimensions such as grammar, structure, and argumentation.

    “A large subset of studies casts GenAI as a grader. During writing and reading assessments, AI delivers analytic diagnoses and scores across dimensions such as grammar, structure, and argumentation; teachers then prune or augment these prompts before returning them to students, who revise line-by-line, after which AI re-checks the latest draft [77,83,110,124,130].”

    From A Systematic Review of Generative AI in K–12: Mapping Goals, Activities, Roles, and Outcomes via the 3P Model (2025) · cites this paper as [124]

    Section 3.5.1, “AI Improving Writing” (page unavailable), describes AI feedback on students’ grammar, language structures, and punctuation. Section 2.1.1 also describes suggestions on grammar and structure. The paper does not establish that a large subset of studies uses GenAI as a grader, or document scores and analytic diagnoses across writing and reading assessments. (LLM, conf=0.87)

    Experimental — accuracy not yet measured

  • These risks have multiple sources: uneven feedback quality, interaction designs that induce excessive cognitive load, and technical constraints (e.g., imperfect speech recognition, misleading 'hallucinations').

    “These risks have multiple sources: uneven feedback quality [124], interaction designs that induce excessive cognitive load [84], and technical constraints (e.g., imperfect speech recognition, misleading 'hallucinations') [78,133].”

    From A Systematic Review of Generative AI in K–12: Mapping Goals, Activities, Roles, and Outcomes via the 3P Model (2025) · cites this paper as [124]

    Section 3.2.3 reports irrelevant information and inconsistent feedback from AI, while the Discussion and Conclusions describes less accurate assessments when AI misinterprets student inputs. These passages support a general risk of misleading output, but the cited paper does not identify imperfect speech recognition or explicitly describe hallucinations as technical constraints. No page numbers were provided in our copy. (LLM, conf=0.82)

    Experimental — accuracy not yet measured

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

Flagged first, then by citations

Research works printed but not linked: 12

  1. Abdelghani, R., Wang, Y.-H., Yuan, X., Wang, T., Lucas, P., Sauzéon, H., & Oudeyer, P.-Y. (2024). GPT-3-driven pedagogical agents to train children's curious question-asking skills. International Journal of Artificial Intelligence in Education, 34(2), 483-518.
  2. Albadarin, Y., Saqr, M., Pope, N., & Turkiainen, M. (2024). A systematic literature review of empirical research on ChatGPT in education. Discover Education, 3, 60.
  3. Black, P., & Wiliam, D. (2009). Developing the theory of formative assessment. Educational Assessment, Evaluation and Accountability, 21(1), 5-31.
  4. Brooks, C., Carroll, A., Gillies, R. M., & Hattie, J. (2019). A matrix of feedback for learning. Australian Journal of Teacher Education, 44, 14-32.
  5. Elstad, E. (2016). Why is there a wedge between the promise of educational technology and the experiences in a technology-rich Pioneer School? In E. Elstad (Ed.), Digital expectations and experiences in education (pp. 77-96). Sense Publishers.
  6. Elstad, E., & Eriksen, H. (2024). High school teachers' adoption of generative AI: Antecedents of instructional AI utility in the early stages of school-specific chatbot implementation. Nordic Journal of Comparative and International Education, 8(1), 1-18.
  7. Farrokhnia, M., Banihashem, S. K., Noroozi, O., & Wals, A. (2023). A SWOT analysis of ChatGPT: Implications for educational practice and research. Innovations in Education and Teaching International, 61(3), 460-474.
  8. Knight, S., Shibani, A., Abel, S., Gibson, A., Ryan, P., Sutton, N., Wight, R., Lucas, C., Sándor, Á., Kitto, K., Liu, M., Vijay Mogarkar, R., & Buckingham Shum, S. (2020). AcaWriter: A learning analytics tool for formative feedback on academic writing. Journal of Writing Research, 12(1), 141-186.
  9. Ma, W., Adesope, O. O., Nesbit, J. C., & Liu, Q. (2014). Intelligent tutoring systems and learning outcomes: A meta-analysis. Journal of Educational Psychology, 106, 901-918.
  10. Shute, V. J., & Rahimi, S. (2017). Review of computer-based assessment for learning in elementary and secondary education. Journal of Computer Assisted Learning, 33(1), 1-19.
  11. Strauss, A., & Corbin, J. (1998). Basics of qualitative research: Techniques and procedures for developing grounded theory. Sage Publications, Inc.
  12. Wiliam, D. (2011). What is assessment for learning? Studies in Educational Evaluation, 37(1), 3-14.

Cited by

Sorted by citations

Related

Authors
Burner, T., Lindvig, Y., & Wærness, J. I.
Affiliations
  • University of South-Eastern Norway (NO)
  • Learnlab, 0164 Oslo, Norway
Corresponding author
Burner, Tony
ORCID
Venue
Education Sciences
Publisher
MDPI AG
Published
2025-01-09
Volume
15
Issue
1
Pages
58
ISSN
2227-7102 (ISSN-L)
DOI
10.3390/educsci15010058
OpenAlex
W4406217983
Cited by
30
References
19
Citations by year
2026: 19 · 2025: 11
FWCI
93.76
Type
journal article
Language
en
Open access
Licence
Locations
  • Education Sciences (publishedVersion, cc-by)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Topics
  • Social Sciences › Education › Student Assessment and Feedback
  • Physical Sciences › Computer Science › Information Systems › Artificial Intelligence in Education
  • Physical Sciences › Computer Science › Artificial Intelligence › Intelligent Tutoring Systems and Adaptive Learning
Keywords
AI-generated feedback, formative assessment, classroom feedback, writing feedback, knowledge base, teacher perceptions, student perceptions, high school students, writing motivation

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review