pith. sign in

arxiv: 2604.16328 · v1 · submitted 2026-03-09 · 💻 cs.CY · cs.SE

Bringing AI into the Classroom: A Structured Approach for Integrating AI into Software Engineering Education

Iris Groher , Michael Vierhauser , Markus Weninger This is my paper

Pith reviewed 2026-05-15 15:31 UTC · model grok-4.3

classification 💻 cs.CY cs.SE
keywords AI-Blueprintssoftware engineering educationAI integrationcurriculum designgenerative AIcomputer science courseseducational resources
0
0 comments X

The pith

AI-Blueprints give educators a structured template to integrate AI topics into software engineering courses.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper introduces AI-Blueprints as a repeatable method for adding AI-related topics and activities to computer science and software engineering curricula. It responds to the rapid rise of generative AI and large language models by supplying concrete guidance where none existed before. The approach includes steps for creating new blueprints and positions them as open educational resources that instructors can adapt. Early evidence comes from interviews with six university educators who reviewed the blueprints for fit with their courses.

Core claim

The paper's central claim is that AI-Blueprints supply a structured process for incorporating AI topics and activities into existing computer science courses, with a defined workflow for generating new blueprints and a plan to release them as open resources for broad adaptation by educators.

What carries the argument

AI-Blueprints: structured templates that specify relevant AI topics, activities, and integration points for a given course or topic.

If this is right

  • Instructors gain a repeatable way to add AI content without redesigning entire courses.
  • Blueprints can be shared and refined as open resources, reducing duplicated effort across institutions.
  • The method supports gradual rather than disruptive curriculum updates as AI tools continue to change.
  • Future expansions of the blueprint library can cover additional topics or course types based on collected feedback.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Widespread use of blueprints could produce more uniform AI skills among graduates entering industry.
  • The templates will require periodic updates to track fast-moving changes in generative AI capabilities.
  • The same blueprint format might transfer to non-CS fields that also need structured AI integration.

Load-bearing premise

Feedback from six semi-structured interviews with university educators is enough to show that the AI-Blueprint method is useful and adaptable across different courses and institutions.

What would settle it

A wider trial in which multiple instructors attempt to adapt the blueprints to their own courses and report that the templates do not fit or require major changes would disprove the central claim.

read the original abstract

The recent emergence of generative AI and Large Language Models (LLMs), particularly following the release of ChatGPT in late 2022, has significantly impacted both academic research and industrial practice. This development has vast potential to impact educational practices across various domains, particularly within computer science and software engineering courses. Unfortunately, there is still a lack of actionable guidance on how to integrate AI technology coherently into computer science curricula. In this paper, we therefore introduce the concept of AI-Blueprints, a structured approach to integrating AI-related topics and activities into various computer science courses. We describe our approach and outline a structured process for creating new blueprints. Our vision is to provide these blueprints as open educational resources, allowing educators to adapt and integrate AI into diverse courses and topics. As a preliminary validation, we conducted semi-structured interviews with six university-level educators, collecting feedback on how our blueprints could help to integrate AI topics into existing courses. Based on this feedback, we lay out plans for future research and expanding our AI-Blueprint concept.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The paper introduces AI-Blueprints as a structured approach for integrating AI-related topics and activities into computer science and software engineering courses. It outlines a process for creating such blueprints, envisions them as open educational resources for educator adaptation, and reports preliminary validation through semi-structured interviews with six university-level educators whose feedback informs plans for future expansion.

Significance. If the AI-Blueprint framework can be shown to support reliable adaptation across varied courses and institutions, it would address a documented gap by supplying concrete, reusable templates for incorporating generative AI and LLMs into existing curricula. The open-resource orientation is a strength that could enable community refinement and wider uptake. The conceptual description of the creation process is clear, but the current evidence base remains too limited to establish these benefits.

major comments (2)
  1. [Preliminary validation] Preliminary validation section: The manuscript presents feedback from six semi-structured interviews as initial support for usefulness and adaptability, yet supplies no information on educator selection criteria, interview protocol, question wording, response coding, thematic analysis, or any negative feedback. Without these details the data cannot credibly substantiate the claim that educators can successfully adapt the blueprints across diverse courses and institutions.
  2. [Abstract and conclusion] Abstract and conclusion: The central assertion that AI-Blueprints constitute a 'structured, adaptable method' rests on a sample of six with no reported quantitative metrics, larger-scale testing, or controlled comparison. This leaves the generalizability claim unsupported at the level required for the paper's stated vision.
minor comments (1)
  1. [Abstract] The abstract would benefit from briefly stating the main themes that emerged from the educator feedback rather than only noting that feedback was collected.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback. We agree that the preliminary validation requires substantially more methodological transparency and that the abstract and conclusion must more accurately reflect the limited scope of the current evidence. We will make the requested revisions in the next version.

read point-by-point responses

  1. Referee: Preliminary validation section: The manuscript presents feedback from six semi-structured interviews as initial support for usefulness and adaptability, yet supplies no information on educator selection criteria, interview protocol, question wording, response coding, thematic analysis, or any negative feedback. Without these details the data cannot credibly substantiate the claim that educators can successfully adapt the blueprints across diverse courses and institutions.

    Authors: We accept this criticism. The original manuscript omitted these details. In revision we will expand the Preliminary Validation section to report: educator selection criteria (including institutional and disciplinary diversity), the complete semi-structured interview protocol with sample questions, the coding and thematic analysis procedure, and a balanced summary that includes any negative or critical comments received from the six participants. These additions will allow readers to evaluate the preliminary nature of the evidence directly. revision: yes

  2. Referee: Abstract and conclusion: The central assertion that AI-Blueprints constitute a 'structured, adaptable method' rests on a sample of six with no reported quantitative metrics, larger-scale testing, or controlled comparison. This leaves the generalizability claim unsupported at the level required for the paper's stated vision.

    Authors: We agree that the present evidence does not support strong generalizability claims. We will revise the abstract and conclusion to (1) explicitly label the interview study as preliminary qualitative validation, (2) remove or qualify language implying broad adaptability across courses and institutions, and (3) foreground the planned future work on larger-scale testing and controlled comparisons. The core contribution remains the description of the AI-Blueprint creation process and the open-resource vision; we will adjust the framing to match the current evidential base. revision: yes

Circularity Check

0 steps flagged

No circularity; AI-Blueprints introduced via external interviews with no self-referential derivation

full rationale

The manuscript presents a conceptual framework (AI-Blueprints) and a process for creating them, validated preliminarily by six semi-structured educator interviews. No equations, fitted parameters, predictions, or first-principles derivations exist. The central claim does not reduce to any input by construction, nor does it rely on self-citation chains or uniqueness theorems. The interviews constitute external data rather than quantities defined inside the paper, satisfying the self-contained criterion. This yields a normal non-finding of circularity.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

This is a conceptual education-framework paper. No free parameters, mathematical axioms, or newly postulated entities are introduced or fitted.

pith-pipeline@v0.9.0 · 5482 in / 1010 out tokens · 52549 ms · 2026-05-15T15:31:28.302464+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

30 extracted references · 30 canonical work pages

  1. [1]

    A. M. Abdelfattah, N. A. Ali, M. A. Elaziz, and H. H. Ammar. Roadmap for Software Engineering Education using ChatGPT. InProc. of the 2023 Int’l Conf. on AI Science and Applications in Industry and Society, pages 1–6. IEEE, 2023

    work page 2023

  2. [2]

    L. W. Anderson and D. R. Krathwohl, editors.A Taxonomy for Learning, Teaching, and Assessing. A Revision of Bloom’s Taxonomy of Educational Objectives. Allyn & Bacon, 2001

    work page 2001

  3. [3]

    Baresi, A

    L. Baresi, A. D. Lucia, A. D. Marco, M. D. Penta, D. D. Ruscio, L. Mariani, D. Micucci, F. Palomba, M. T. Rossi, and F. Zampetti. Students’ Perception of ChatGPT in Software Engineering: Lessons Learned from Five Courses. InProc. of the 37th Intl’ Conf. on Softw. Eng. Education and Training. IEEE, 2025

    work page 2025

  4. [4]

    Bearman, J

    M. Bearman, J. Ryan, and R. Ajjawi. Discourses of artificial intelligence in higher education: A critical literature review.Higher Education, 86(2):369–385, 2023

    work page 2023

  5. [5]

    J. Biggs. Enhancing teaching through constructive alignment.Higher education, 32(3):347–364, 1996

    work page 1996

  6. [6]

    Cambaz and X

    D. Cambaz and X. Zhang. Use of AI-driven code generation models in teaching and learning programming: a systematic literature review. InProc. of the 55th ACM Technical Symp. on Computer Science Education, pages 172–178. ACM, 2024

    work page 2024

  7. [7]

    Charmaz.Constructing grounded theory: A practical guide through qualitative analysis

    K. Charmaz.Constructing grounded theory: A practical guide through qualitative analysis. Sage, 2006

    work page 2006

  8. [8]

    Cockburn

    A. Cockburn. Basic Use Case Template.Humans and Technology, Technical Report, 96:28, 1998

    work page 1998

  9. [9]

    Dickey, A

    E. Dickey, A. Bejarano, and C. Garg. AI-Lab: A Framework for Introducing Generative Artificial Intelligence Tools in Computer Programming Courses.SN Computer Science, 5(6):720, 2024

    work page 2024

  10. [10]

    A. S. Fernandez, D. Patrick, M. Gomez, and K. A. Cornell. Incorporating llm activities into established cs1 curriculum: An experience report.Journal of Computing Sciences in Colleges, 40(8):79–93, 2025

    work page 2025

  11. [11]

    Frankford, C

    E. Frankford, C. Sauerwein, P. Bassner, S. Krusche, and R. Breu. AI-Tutoring in Software Engineering Education. InProc. of the 46th Int’l Conf. on Softw. Eng.: SEET, pages 309–319. ACM, 2024

    work page 2024

  12. [12]

    Greer, Q

    T. Greer, Q. Hao, M. Jing, and B. Barnes. On the effects of active learning environments in computing education. InProc. of the 50th ACM Technical Symposium on Computer Science Education, page 267–272. ACM, 2019

    work page 2019

  13. [13]

    A. Guha, B. Zorn, C. J. Anderson, M. Q. Feldman, S. Gulwani, and G. Allen. The future of programming in the age of large language models.Future, 2025

    work page 2025

  14. [14]

    H. C. Herring and J. R. Williams. The role of objectives in curriculum development. Journal of Accounting Education, 18(1):1–14, 2000

    work page 2000

  15. [15]

    X. Hou, Y. Zhao, Y. Liu, Z. Yang, K. Wang, L. Li, X. Luo, D. Lo, J. Grundy, and H. Wang. Large Language Models for Software Engineering: A Systematic Literature Review. ACM Trans. on Soft. Eng. and Methodology, 33(8):1–79, 2024

    work page 2024

  16. [16]

    Kasneci, K

    E. Kasneci, K. Seßler, S. Küchemann, M. Bannert, D. Dementieva, F. Fischer, U. Gasser, G. Groh, S. Günnemann, E. Hüllermeier, et al. ChatGPT for good? On opportunities and challenges of large language models for education.Learning and individual differences, 103:102274, 2023

    work page 2023

  17. [17]

    V. D. Kirova, C. S. Ku, J. R. Laracy, and T. J. Marlowe. Software Engineering Education Must Adapt and Evolve for an LLM Environment. InProc. of the 55th ACM Technical Symposium on Computer Science Education, pages 666–672. ACM, 2024

    work page 2024

  18. [18]

    A. N. Kumar and R. K. Raj. Computer science curricula 2023 (cs2023): The final report. InProc. of the 55th ACM Technical Symposium on Computer Science Education V. 2, page 1867–1868. ACM, 2024

    work page 2023

  19. [19]

    Y. Li, J. Keung, and X. Ma. Integrating Generative AI in Software Engineering Education: Practical Strategies. InProc. of the 2024 International Symposium on Educational Technology, pages 49–53. IEEE, 2024

    work page 2024

  20. [20]

    J. Luo, C. Zheng, J. Yin, and H. H. Teo. Design and assessment of ai-based learning tools in higher education: a systematic review.International Journal of Educational Technology in Higher Education, 22(1):42, 2025

    work page 2025

  21. [21]

    Malhotra, M

    R. Malhotra, M. Massoudi, and R. Jindal. Shifting from traditional engineering educa- tion towards competency-based approach: The most recommended approach-review. Education and Information Technologies, 28(7):9081–9111, 2023

    work page 2023

  22. [22]

    C. K. Sah, L. Xiaoli, M. M. Islam, and M. K. Islam. Navigating the AI Frontier: A Critical Literature Review on Integrating Artificial Intelligence into Software Engineering Education. InProc. of the 36th Int’l Conf. on Softw. Eng. Education and Training, pages 1–5. IEEE, 2024

    work page 2024

  23. [23]

    D. A. Schmidt, B. Alboloushi, A. Thomas, and R. Magalhaes. Integrating artificial intelligence in higher education: perceptions, challenges, and strategies for academic innovation.Computers and Education Open, 9:100274, 2025

    work page 2025

  24. [24]

    Sengul, R

    C. Sengul, R. Neykova, and G. Destefanis. Software engineering education in the era of conversational AI: current trends and future directions.Frontiers in AI, 7, 2024

    work page 2024

  25. [25]

    Sharifani and M

    K. Sharifani and M. Amini. Machine learning and deep learning: A review of methods and applications.World Information Technology and Engineering Journal, 10(07):3897– 3904, 2023

    work page 2023

  26. [26]

    Sheese, M

    B. Sheese, M. Liffiton, J. Savelka, and P. Denny. Patterns of student help-seeking when using a large language model-powered programming assistant. InProc. of the 26th Australasian Computing Education Conf., pages 49–57. ACM, 2024

    work page 2024

  27. [27]

    Skublewska-Paszkowska, M

    M. Skublewska-Paszkowska, M. Miłosz, and E. Lukasik. Acm/ieee recommendations for computing curricula and the needs of the polish cs industry. InProc. of the 9th Int’l Conf. on Education and New Learning Technologies, pages 9050–9057. IATED, 2017

    work page 2017

  28. [28]

    C. W. Starr, B. Manaris, and R. H. Stalvey. Bloom’s taxonomy revisited: specifying assessable learning objectives in computer science.ACM Sigcse Bulletin, 40(1):261–265, 2008

    work page 2008

  29. [29]

    Vierhauser, I

    M. Vierhauser, I. Groher, T. Antensteiner, and C. Sauerwein. Towards integrating emerging AI applications in SE education. InProc. 36th Int’l Conf. on Softw. Eng. Education and Training, pages 1–5. IEEE, 2024

    work page 2024

  30. [30]

    Yabaku and S

    M. Yabaku and S. Ouhbi. University Students’ Perception and Expectations of Genera- tive AI Tools for Software Engineering. InProc. of the 36th Int’l Conf. on Softw. Eng. Education and Training, pages 1–5. IEEE, 2024. 6

    work page 2024