Perceived Localized Guidelines Used to Distinguish Between Human-Written and AI-Generated Texts

Abstract

This research examined the perceived localized guidelines used by teachers to evaluate the authenticity of written texts, particularly in distinguishing between human-written and AI-generated outputs among Intermediate Elementary Teachers of Talamban Elementary School, North District 8, Cebu City Division, Talamban, Cebu, Philippines. Specifically, it determined the teachers’ profile in terms of age, sex, years of teaching experience, subject area taught, and highest educational attainment; assessed the teachers’ level of awareness in terms of accuracy in distinguishing human-written and AI-generated texts; identified the indicators teachers used in making authorship judgments (e.g., vocabulary level, sentence structure, grammar, repetition and patterns, creativity and originality and coherence and flow), and described the issues and concerns encountered in identifying AI-generated text. A descriptive–correlational research design was employed using a validated structured questionnaire administered to the teacher-respondents. The descriptive component was used to summarize respondents’ profiles and their reported awareness in terms of accuracy, indicators and issues and concerns, while the correlational component was used to determine whether a significant relationship between teachers\' level of awareness in terms of accuracy and the indicators used in evaluating the authenticity of written texts. The findings of the study served as the basis for developing perceived localized guidelines to distinguish between AI-generated and human written texts, thereby strengthening teachers’ capacity to evaluate the authenticity of students’ written outputs.

Introduction

The study investigates teachers’ ability to distinguish human-written English text from AI-generated text among Key Stage 2 learners at Talamban Elementary School, North District 8, Talamban, Cebu City, during School Year 2025–2026. It focuses on the growing challenge of maintaining academic integrity, originality, and authentic assessment as generative AI becomes increasingly capable of producing fluent and human-like writing.

The study argues that although AI-detection software exists, such tools can produce inaccurate or inconsistent results. Therefore, teachers’ professional judgment, contextual knowledge, and evaluation of students’ writing remain important. Teachers may consider factors such as vocabulary, sentence structure, grammar, repetition, creativity, originality, coherence, flow, personal voice, and consistency with a student's previous work.

Research Problem and Objectives

The research aims to determine teachers’ ability to accurately identify human-written and AI-generated text. Specifically, it examines:

• Teachers’ demographic characteristics, including age, sex, teaching experience, subject area, and educational attainment.
• Their accuracy in identifying human and AI-generated writing.
• The indicators they use when judging writing authenticity.
• The relationship between teachers’ awareness and their identification accuracy.
• The issues and concerns teachers face when distinguishing AI-generated writing.
• The development of localized guidelines for identifying and addressing potentially AI-generated student work.

The proposed guidelines are intended to help teachers improve assessment practices, document and verify suspected cases fairly, promote responsible AI use, and assist school administrators in developing appropriate policies.

Literature Review

The literature shows that AI-generated writing can imitate human fluency, structure, coherence, and academic tone, making authorship increasingly difficult to determine. Research also indicates that teachers cannot always identify AI-generated writing accurately and may be influenced by the quality of the AI output, the writing task, and their own expectations.

Research on contract cheating and academic integrity similarly shows that teachers often rely on professional intuition, previous student performance, writing style, and contextual evidence, but these indicators are not always reliable. The literature therefore emphasizes that AI detection should not depend exclusively on automated software or isolated textual features.

Instead, teachers should consider multiple forms of evidence, including:

• Writing style and personal voice
• Vocabulary and sentence patterns
• Coherence and depth of ideas
• Student's previous writing
• Task requirements and conditions
• Evidence of the student's learning process
• Consistency between the submitted work and the student's demonstrated abilities

Overall, the literature supports a human-centered approach in which technology assists teachers but does not replace professional judgment.

Methodology

The study employed a descriptive-correlational research design. It involved 30 teachers from Talamban Elementary School, selected through purposive sampling.

The researcher-developed questionnaire contained four major parts:

1. Respondent Profile – demographic and professional characteristics.
2. Awareness and Accuracy – teachers’ ability to distinguish AI-generated from human-written text.
3. Authenticity Indicators – factors teachers consider when evaluating written work.
4. Issues and Concerns – difficulties experienced in identifying AI-generated writing.

A five-point scale was used for the issues and concerns section. The study employed:

• Frequency and percentage for demographic characteristics.
• Weighted mean for awareness, authenticity indicators, and issues.
• Spearman's rho to determine the relationship between teachers’ awareness and classification accuracy.

Respondent Profile

The study found that:

• The largest age group was 31–40 years, followed by teachers aged 51 and above, although the estimated overall mean age was 40.5 years with a standard deviation of 11.37.
• All 30 respondents were female (100%).
• 40% had 16 or more years of teaching experience, making veteran teachers the largest experience group.
• English and Filipino teachers each represented 20% of the sample, while Math, Science, and Araling Panlipunan each represented 13.33%.
• 66.67% had units in a Master's degree, while 6.67% had completed a Master's degree and 26.66% held a Bachelor's degree.
• No respondents had doctoral units or a doctoral degree.

The respondent profile therefore represents teachers with varied ages, teaching experience, subject backgrounds, and substantial exposure to graduate-level professional development.

Main Significance

The central idea of the research is that AI-generated writing cannot always be reliably identified through technology or surface-level writing characteristics alone. Teachers need a combination of professional experience, contextual knowledge, assessment evidence, and carefully selected indicators.

The study ultimately aims to develop localized guidelines for Talamban Elementary School that can help teachers distinguish suspected AI-generated work more systematically and fairly. Such guidelines should avoid automatically treating polished or unusual writing as evidence of AI use and should instead encourage verification through student writing history, oral questioning, process-based assessment, and other contextual evidence.

Conclusion

Based on the findings, the study concluded that teachers at Talamban Elementary School, North District 8, Cebu City Division demonstrated a high level of awareness of both human-written and AI-generated texts. They utilized multiple indicators, particularly grammar, vocabulary, coherence and flow, and originality, in evaluating the authenticity of written outputs. The study further revealed a significant relationship between teachers’ awareness and their use of indicators in assessing text authenticity, suggesting that greater awareness is associated with more effective evaluation practices. In addition, teachers experienced moderate concerns regarding the use of AI-generated texts in educational settings. These findings support the need to develop and implement localized guidelines to assist teachers in addressing AI-related issues and maintaining authenticity in writing assessments.

References

A. Books and Journal Articles [1] Fleckenstein, J., Meyer, J., Jansen, T., & Keller, S. (2024). Do teachers spot AI? Evaluating the detectability of AI-generated texts among student essays. *Computers and Education: Artificial Intelligence, 6*, 100209. https://doi.org/10.1016/j.caeai.2024.100209 [2] Green, D. M., & Swets, J. A. (1966). *Signal detection theory and psychophysics*. John Wiley & Sons. [3] King, P. M., & Kitchener, K. S. (1994). *Developing reflective judgment: Understanding and promoting intellectual growth and critical thinking in adolescents and adults*. Jossey-Bass. [4] Sadler, D. R. (1989). Formative assessment and the design of instructional systems. *Instructional Science, 18*(2), 119–144. https://doi.org/10.1007/BF00117714 [5] Sadler, D. R. (2010). Beyond feedback: Developing student capability in complex appraisal. *Assessment & Evaluation in Higher Education, 35*(5), 535–550. https://doi.org/10.1080/02602930903541015 [6] Wiggins, G. (1998). *Educative assessment: Designing assessments to inform and improve student performance*. Jossey-Bass. [7] Fleckenstein, J., Meyer, J., Jansen, T., & Keller, S. (2024). Do teachers spot AI? Evaluating the detectability of AI-generated texts among student essays. Computers and Education: Artificial Intelligence, 6, 100209. https://doi.org/10.1016/j.caeai.2024.100209 [8] Dawson, P., & Sutherland-Smith, W. (2018). Can markers detect contract cheating? Results from a pilot study. Assessment & Evaluation in Higher Education, 43(2), 286–293. https://doi.org/10.1080/02602938.2017.1336746 [9] Eaton, S. E., & Dressler, R. (2021). Plagiarism in higher education: Tackling tough topics in academic integrity. Springer. https://doi.org/10.1007/978-3-030-63379-3 [10] Lancaster, T. (2020). Academic integrity and contract cheating: Responses from the UK and beyond. International Journal for Educational Integrity, 16, 16. https://doi.org/10.1007/s40979-020-00070-0 [11] Lo, C. K. (2023). Artificial intelligence and academic integrity: Student and teacher perspectives. Computers & Education, 201, 104788. https://doi.org/10.1016/j.compedu.2023.104788 [12] Simpson, A. V. (2020). Detecting plagiarism and contract cheating: The role of assessment design. Assessment & Evaluation in Higher Education, 45(5), 749–761. https://doi.org/10.1080/02602938.2019.1683362 [13] Stoesz, B. M., Eaton, S. E., Miron, J. B., & Thacker, E. J. (2019). Academic integrity and contract cheating: Faculty views and detection strategies. Journal of Academic Ethics, 17(3), 245–269. https://doi.org/10.1007/s10805-019-09335-7 B. Legal Documents, Republic Acts, and DepEd Orders [1] Board for Professional Teachers. (1997). *Resolution No. 435, s. 1997: Code of Ethics for Professional Teachers*. Professional Regulation Commission, Republic of the Philippines. [2] Congress of the Philippines. (1989). *Republic Act No. 6713: Code of Conduct and Ethical Standards for Public Officials and Employees*. Official Gazette of the Republic of the Philippines. https://www.officialgazette.gov.ph/1989/02/20/republic-act-no-6713/ [3] Congress of the Philippines. (1994). *Republic Act No. 7836: Philippine Teachers Professionalization Act of 1994*. Official Gazette of the Republic of the Philippines. https://www.officialgazette.gov.ph/1994/12/16/republic-act-no-7836/ [4] Congress of the Philippines. (1997). *Republic Act No. 8293: Intellectual Property Code of the Philippines*. Official Gazette of the Republic of the Philippines. https://www.officialgazette.gov.ph/1997/06/06/republic-act-no-8293/ [5] Congress of the Philippines. (2004). *Republic Act No. 9293: An Act Amending Certain Sections of Republic Act No. 7836*. Official Gazette of the Republic of the Philippines. https://www.officialgazette.gov.ph/2004/04/21/republic-act-no-9293/ [6] Congress of the Philippines. (2012). *Republic Act No. 10173: Data Privacy Act of 2012*. Official Gazette of the Republic of the Philippines. https://www.officialgazette.gov.ph/2012/08/15/republic-act-no-10173/ [7] Department of Education. (2017). *DepEd Order No. 42, s. 2017: National Adoption and Implementation of the Philippine Professional Standards for Teachers*. Department of Education, Republic of the Philippines. [8] Official Gazette of the Republic of the Philippines. (1987). *The 1987 Constitution of the Republic of the Philippines – Article XIV: Education, Science and Technology, Arts, Culture and Sports*. https://www.officialgazette.gov.ph/constitutions/the-1987-constitution-of-the-republic-of-the-philippines/the-1987-constitution-of-the-republic-of-the-philippines-article-xiv/

Copyright

Copyright © 2026 Miraflor A. Lecciones. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.