Teacher's Guide to AI

 

A New Pedagogy for the Age of Generative AI: A Teacher's Guide to Structured Integration and Authentic Assessment

(Download the PDF for review later here)

Executive Summary

The rapid and widespread adoption of generative artificial intelligence (GenAI) tools by students has fundamentally reshaped the educational landscape. Large, recent surveys indicate that GenAI use for coursework is now mainstream, with a dramatic increase in student usage observed from 2024 to 2025. This report argues that a simple, defensive posture—characterized by bans and reliance on flawed detection tools—is neither viable nor pedagogically sound. The research from leading educational bodies, including the Modern Language Association-Conference on College Composition and Communication (MLA-CCCC), Jisc, and EDUCAUSE, provides a clear path forward.

This report synthesizes current research and guidance into a complete, actionable framework for educators. The central thesis is that the challenge of AI is, in fact, an opportunity to cultivate a more robust, process-oriented, and authentic model of learning. The recommendations are grounded in three core principles:

1. Abandon Detection, Embrace Process: Given the unreliability and fairness limitations of AI detectors, particularly for non-native English speakers, assessment must pivot from policing a final product to evaluating the transparent, human-driven process of creation.

2. Teach AI Literacy: The effective and ethical use of GenAI is an essential skill for the modern world. Educators are urged to explicitly teach students how to use, critique, audit, and responsibly cite AI tools.

3. Redesign for Authenticity: Assignments should be re-engineered to foreground human reasoning, situated knowledge, and metacognition. This can be achieved through innovative pedagogical approaches such as rhetorical prompt design, interactive oral defenses, and provenance-based portfolios.

The report provides a detailed guide for implementing these principles, including a clear matrix for creating course policies and six ready-to-run assignment patterns that can be adapted for any discipline. By embracing this proactive approach, educators can transform a potential threat to academic integrity into a powerful catalyst for deeper, more meaningful learning.

I. The Inevitable Integration: Understanding the Current Landscape

1.1 The New Normal: GenAI Usage in Higher Education

Generative AI is no longer a niche technology; it has become a mainstream utility within higher education. Recent and large-scale surveys, such as those conducted by HEPI and the Financial Times, reveal that a majority of undergraduates now utilize GenAI for their coursework. This use is often for foundational tasks like acting as a tutor or drafting aid. The most significant finding is the dramatic increase in adoption, with usage jumping sharply from 2024 to 2025 [User Query]. This rapid acceleration marks a critical inflection point, indicating that the technology has moved from an experimental tool for a few to a ubiquitous resource for many.

The widespread availability and accessibility of these tools mean that the traditional pedagogical model, which assumes that students engage with assignments without such aids, is now functionally obsolete. The era of simply prohibiting these tools has passed, as such policies are increasingly difficult to enforce and fail to address the core reality of student behavior. The shift in the landscape compels educators to move beyond a defensive posture of prohibition and policing. The central question for the teaching community is no longer whether students will use AI, but rather how they can be guided to use it productively, ethically, and in a manner that supports, rather than subverts, their learning. The focus must now shift from managing a threat to integrating a powerful new tool.

1.2 The Learning Effects of AI: Mixed Outcomes, Promising Horizons

The impact of GenAI on student learning is not uniform; it is profoundly influenced by the pedagogical context in which it is used. Research from institutions like Heinz College, along with studies cited in journals like PMC and ScienceDirect, presents a nuanced picture: when the use of AI is structured and guided, the learning effects are promising [User Query]. For example, controlled studies from 2024–2025 found that AI feedback and support can significantly improve writing quality and efficiency, a benefit particularly notable for English as a Second Language (ESL) students [User Query]. This finding illustrates a powerful potential for AI to act as a personalized and accessible learning aid.

Conversely, unstructured, "just use ChatGPT" approaches, where students are left to their own devices without clear guidance, often lead to shallow engagement [User Query]. This dichotomy suggests that the promise of GenAI's effectiveness is not inherent in the technology itself but is directly tied to the quality of the pedagogical framework established by the educator. In this context, the role of the teacher is not diminished but transformed. The educator becomes a strategic architect of learning experiences, designing assignments and systems that harness the tool's power for specific, productive ends. The goal is to move students from being passive consumers of AI-generated content to active, critical, and strategic users. This evolution of the teacher's role is foundational to a forward-looking pedagogical approach.

II. The Paradigm Shift: Moving Beyond Detection

2.1 The Limits of AI Detectors: Why Detection Is Not a Pedagogy

In the initial response to the rise of GenAI, many institutions and educators turned to AI detection software as a primary defense against academic misconduct. However, a significant body of research and guidance from authoritative sources has exposed the deep-seated limitations of these tools [User Query]. Research and policy papers from institutions like Stanford HAI and documents archived on platforms like arXiv highlight that AI detectors have considerable reliability and fairness issues.1 A particularly concerning problem is the risk of false positives, which disproportionately affect non-native writers [User Query]. This vulnerability creates a risk of wrongly accusing students, which can have severe and unfair consequences.

Major organizations are now actively advising against the sole reliance on these detectors as a definitive measure of academic integrity. The MLA-CCCC, for instance, has published guidance that cautions educators against using such software due to its known biases and lack of accuracy.3 The reliance on detection tools also cultivates an adversarial dynamic in the classroom. Instead of a collaborative environment focused on learning, the classroom becomes a space of surveillance and suspicion. This approach not only fails to accurately address the issue of GenAI use but also actively undermines the trust and transparency necessary for fostering genuine intellectual engagement. By closing the door on the flawed premise of detection, educators can open a new one to more effective and equitable pedagogical strategies.

2.2 A New Philosophy of Assessment: Design for Authenticity, Transparency, and Process

The advent of GenAI compels a fundamental re-evaluation of assessment itself. National bodies such as the MLA-CCCC, Jisc, QAA, and EDUCAUSE have coalesced around a new philosophy: assessment must be rethought, not abandoned.4 This collective guidance recommends designing assignments that are characterized by transparency, process, authenticity, and AI literacy.7 The widespread accessibility of GenAI forces educators to question a foundational assumption of traditional academia: that an assessment output is, by itself, evidence of a meaningful learning process. The European University Association (EUA), in partnership with QAA, notes that AI-generated outputs can not only fool assessors but may also be marked higher than human-written work, directly challenging this assumption.8

This situation has profound implications for academic standards. If AI-assisted outputs lead to a gradual inflation of grades, the very value and comparability of academic awards could be undermined, creating a potential public crisis of confidence in the degree itself.8 To avert this, educational institutions must adopt a broadly consistent approach based on shared principles rather than an inflexible, top-down set of rules.8 This means that the focus of assessment must shift from a final, static product to a dynamic, transparent process. By designing for authenticity—tasks that mirror real-world applications—and for process—evaluating the steps a student takes to arrive at a conclusion—educators can reclaim the true purpose of assessment: to measure what a student has genuinely learned and can do.

2.3 Crafting Explicit AI Policies for Your Course

A crucial step in this new pedagogical approach is the creation of clear, explicit AI policies that are communicated directly in the syllabus and on assignment briefs. This is a deliberate move away from vague, ambiguous statements or outright bans. Leading institutions like UCLA and Harvard's Bok Center are providing templates and guidance for this purpose, with resources available through their teaching and learning centers.6 These models advocate for a per-assignment framework, similar to the Monash or COMPASS styles, which clearly signals the permitted use of AI for each specific task [User Query].

This level of detail is not merely a bureaucratic exercise; it is a fundamental act of pedagogical design. By defining exactly how AI can and cannot be used for each assignment, the instructor is forced to articulate their learning objectives with greater precision. This transparency signals to students that the course is not a "trap," but a transparent environment with clear rules of engagement for a powerful new tool. It builds a foundation of trust and reduces the incentive for students to conceal their use of AI.

The following matrix provides a clear, actionable template for this approach, with specific examples and disclosure requirements for each policy framework.

Policy Framework	Definition	Permitted Uses	Forbidden Uses	Disclosure Requirements
AI-Free	This policy is for assignments designed to measure foundational, human-driven skills. The integrity of the task depends on the student's unassisted work.	No use of GenAI tools (e.g., ChatGPT, Claude, Midjourney) is permitted at any stage of the assignment.	Using AI to brainstorm ideas, draft content, or edit grammar.	No disclosure required.
AI-Assisted	This policy acknowledges GenAI as a utility for efficiency and support. The core intellectual work must remain with the student.	Using AI for tasks that do not generate the core content or argument, such as brainstorming, refining search queries, or checking for grammatical errors.	Using AI to write full sections, generate arguments, or create the final product.	A short, sentence-long statement detailing the use.
AI-Integrated	This policy frames GenAI as a required tool for completing the assignment. The task's learning outcomes depend on the student's strategic and critical use of AI.	Using AI to generate initial drafts, comparative models, or data visualizations. The student's primary task is to critically analyze, revise, and verify the AI output.	Submitting AI output without significant human revision or verification.	A detailed statement explaining what AI was used, why, and how the output was verified.
AI-Central	This policy makes AI the central subject of the assignment. The learning outcome is to explore, critique, and audit the AI itself.	Any use of AI is permitted and required to complete the assignment's goals.	The student failing to provide the full provenance of their work (e.g., all prompts, chat excerpts, and an audit of the output).	A comprehensive appendix or portfolio that includes all AI prompts, chat excerpts, and a reflective analysis of the process.

---

III. Foundational Principles for AI-Integrated Pedagogy

3.1 Cultivating AI Literacy as a Core Learning Outcome

Rather than simply imposing rules on AI use, a more durable and effective strategy is to teach AI literacy as a core learning outcome. The MLA-CCCC, in its "Student Guide to AI Literacy," frames this as an essential skill for navigating the modern world.12 The guide outlines key competencies that go far beyond simple tool usage.13

A comprehensive approach to AI literacy includes:

• Foundational Understanding: Students should learn how GenAI systems work at a basic level, understanding that large language models (LLMs) are essentially predicting the most likely next word rather than understanding concepts.13 They should also recognize that these systems are influenced by human intervention, including feedback and content moderation.13

• Effective Prompting: The act of prompting is a skill in itself. The EDUCAUSE "CLEAR" framework—Concise, Logical, Explicit, Adaptive, and Reflective—offers a mindset for crafting effective prompts.14 This goes beyond simply copying prompts from a library; it requires a strategic and articulate approach to getting the desired output from the model.

• Critical Evaluation: A literate user understands that AI outputs are not inherently correct or unbiased. Students must be taught to check the accuracy and relevance of AI output against credible, external sources.13 The practice of identifying and correcting "hallucinations" or biased language in AI-generated text reinforces the critical thinking and fact-checking skills fundamental to research and scholarship.

• Transparent Use: Students must learn to follow relevant guidelines, properly credit GenAI contributions, and be prepared to discuss their process openly with their instructors and peers.13 This transparency transforms the academic-integrity conversation from a punitive one into an educational one.

By teaching these skills, educators are not merely responding to a new technology; they are preparing students to be more discerning consumers and creators of information in a rapidly evolving digital ecosystem. This practice of auditing and critiquing AI empowers students with a meta-skill that extends far beyond the AI context.

3.2 Rhetorical Prompt Design: A New Frontier of Composition

The act of writing has always been a form of thinking, and the introduction of GenAI does not diminish this central premise. Instead, it reconfigures the process, elevating the importance of rhetorical choice. The WAC Clearinghouse's TextGenEd collection showcases how educators are now framing "prompting" as a new form of composition, or "rhetorical prompt engineering".15

In this pedagogical approach, the intellectual labor is no longer focused on the rote generation of text, but on the strategic decision-making that precedes it. Students are tasked with iterating on prompts and justifying their choices using rhetorical vocabulary, such as audience, purpose, and genre.15 The final output may be generated by an AI, but the student remains the creative agent, responsible for the high-level intellectual work of crafting the intent and constraints that guide the AI. For instance, a student might generate a series of drafts with an AI—one for a professional audience, another for a general readership, and a third in a different genre—and then reflect on how their prompt edits changed the output and why. This process positions AI not as a replacement for thinking, but as a powerful tool for prototyping ideas and exploring rhetorical options, thereby preserving and even enhancing the core goals of a writing course.

3.3 The Power of Provenance: Process Portfolios and Learning Journals

As assessment shifts from a static product to a dynamic process, portfolios become an invaluable tool. National guidance from Jisc, QAA, and EDUCAUSE, along with the MLA-CCCC, endorses the use of provenance-based portfolios as a way to assess learning in the age of AI.7 These portfolios collect and document the entire journey of a project, not just the final outcome. The evidence of a student's process can include ideation notes, multiple drafts with version history, excerpts from AI chat logs, and a short AI use statement explaining what was used, why, and how it was verified.3

This approach is not simply a defensive measure against cheating; it is a superior pedagogical practice. By focusing on the journey, it provides a richer, more accurate picture of a student's learning and reduces the incentive for students to engage in contract cheating or misuse AI. An AI-use statement, for example, compels students to reflect on the impact of the tool on their work, transforming what might have been a clandestine act into an explicit and educational part of the assignment. The portfolio's collection of drafts and feedback provides concrete evidence of a student's growth and effort over time. The instructor can use this body of evidence to have a productive dialogue with the student about their learning, rather than a confrontational one about potential misconduct. As the MLA-CCCC blog notes, this process-based approach is a non-adversarial alternative to unreliable AI text detectors.3

IV. The New Classroom: Ready-to-Run Assignment Patterns

This section provides six actionable assignment patterns that can be adapted for a wide range of disciplines and learning objectives. These examples move beyond theory and provide concrete, ready-to-run models for integrating GenAI in a structured, pedagogically sound way. The following table provides a quick overview of each pattern.

Assignment Pattern	Primary Pedagogical Goal	Key Components	Ideal Use Case
AI-Idea to Human-Argument	Fostering research and argumentation skills.	AI-generated topic map, human-designed research plan, evidence-based essay, AI use statement.	Introductory writing courses, research methods classes.
Rhetorical Prompting Log	Enhancing rhetorical awareness and critical analysis.	A series of prompts, annotated log of changes, reflective memo.	Composition courses, communications, rhetoric.
Model-vs-Human Audit	Cultivating fact-checking and critical editing.	A flawed AI-generated draft, library research, student revisions, change log.	Research-intensive seminars, technical writing, journalism.
Interactive Oral Defense (IOA)	Assessing reasoning, situated knowledge, and public speaking.	A written artifact (with disclosure), recorded dialogic interview.	Any discipline where reasoning and transfer of knowledge are key.
Source-Triangulation + AI	Building synthesis and research literacy.	AI-generated claims, peer-reviewed sources, synthesis essay, methods appendix.	Upper-level humanities or social science seminars.
Provenance Portfolio	Promoting metacognition and process-oriented learning.	Collection of drafts, notes, feedback, citations, and reflections over a term.	Capstone projects, e-portfolios, semester-long courses.

---

4.1 Pattern 1: AI-Idea to Human-Argument

This assignment re-engineers the traditional research paper by segmenting the tasks to ensure that the core intellectual work remains with the student. The goal is to use AI at the beginning of the process for a limited, well-defined purpose, thereby removing the burden of finding an initial topic while forcing the student to perform the more complex tasks of research and argumentation.3

Assignment Steps:

1. AI-Generated Topic Map: Students use a GenAI tool to generate a topic map or an outline of a controversial subject. For example, a student might prompt the AI to "create a topic map of the key debates surrounding the ethics of self-driving cars." The AI's output is treated as a starting point, not a final answer.

2. Human-Designed Research Plan: Based on the AI's topic map, the student develops a research plan. This plan identifies specific questions to investigate and outlines a strategy for locating credible, human-authored sources from the library.

3. Evidence-Based Essay: The student then writes a claim-and-evidence essay, building a nuanced argument based solely on their human-sourced research. The essay is graded on the quality of its argumentation and the strength of its evidence, not the AI output.

4. Provenance Documentation: Students attach the initial AI-generated topic map and a brief, 150-word AI use statement explaining how the AI was used, how it contributed to their process, and the specific guardrails they employed.

4.2 Pattern 2: Rhetorical Prompting Log

This assignment teaches the art and science of rhetorical prompt design, framing the interaction with GenAI as a new form of composition. Students are required to be metacognitive about their use of the tool, documenting and reflecting on their choices.

Assignment Steps:

1. Develop a Core Claim: Students first establish a single, clear claim.

2. Iterate on Prompts: The student then writes 4–6 different prompts to get an AI model to write on that same claim but for different audiences, purposes, or genres. For instance, one prompt might ask for an academic abstract, another for a blog post, and a third for a social media caption.

3. Annotate the Log: The student records each prompt and the resulting AI output in a log. They annotate the log, explaining how each change in the prompt—a new constraint, a different tone, or a specific persona—affected the AI's output.

4. Reflective Memo: The assignment culminates in a reflective memo. In this memo, students analyze the constraints and affordances of the AI tool, discussing its strengths and limitations for different rhetorical tasks. The memo is the final, graded product, not the AI's output.14

4.3 Pattern 3: Model-vs-Human Audit

This pattern shifts the student's role from a writer to a critical editor and auditor. The assignment begins with a flawed, pre-written AI draft, and the student's task is to improve it. This directly addresses the need for students to fact-check and critically evaluate AI outputs.

Assignment Steps:

1. Instructor-Provided Draft: The instructor provides a GenAI-generated draft that contains factual errors, logical inconsistencies, or weak argumentation (e.g., "hallucinations") [User Query].

2. Verification and Revision: Students are tasked with two primary objectives:

• Verify Factual Claims: Students must use library and peer-reviewed sources to verify or refute every factual claim in the provided draft.

• Rewrite Shaky Sections: They must rewrite any sections that are factually inaccurate, poorly argued, or otherwise weak.

3. Change Log: The final product includes a "change log" where the student explains every fix they made and provides a rationale for each revision. The rubric for the assignment heavily weighs the quality of the verification and the soundness of the revision rationale, rather than the final rewritten text itself.

4.4 Pattern 4: Interactive Oral Defense (IOA)

The Interactive Oral Defense (IOA), or viva-style assessment, is a powerful tool for foregrounding reasoning, transfer, and situated knowledge, making it particularly effective in the age of GenAI. The written artifact itself, which may be GenAI-assisted, becomes a starting point for a dialogic interview where the student defends their work [User Query].

Assignment Steps:

1. Artifact Submission: Students submit a 1–2 page written brief or artifact, with an explicit disclosure of any AI use.

2. Recorded Q&A: The student then participates in a recorded, 10-minute interview with the instructor or a small group of peers. In this interview, they are asked to defend the choices they made in their work, discuss the limitations of their research, and articulate the next steps for the project [User Query].

3. Rubric Focus: The assessment is based on a simple rubric that evaluates the student's reasoning, application of concepts, and responsiveness to questioning. The University of Sydney has successfully piloted this method in large first-year units, with students taking on roles like "writers pitching their work to an editor" and using AI chatbots trained on course content to practice for the interview beforehand.16

4.5 Pattern 5: Source-Triangulation + AI

This assignment uses GenAI as a prompt generator to create a scaffolded research task that requires students to engage deeply with credible sources. It leverages the AI's ability to generate multiple viewpoints on a topic, while ensuring the student's learning is focused on the human skills of synthesis and critical evaluation [User Query].

Assignment Steps:

1. AI-Generated Claims: Students use an AI to generate three opposing claims or viewpoints on a chosen controversy. For example, the student could ask, "What are three opposing claims about the long-term economic effects of remote work?"

2. Human-Driven Research: The student's core task is to then find peer-reviewed and credible sources to confirm or refute each of the AI-generated claims. This requires genuine information literacy and source-evaluation skills.

3. Synthesis and Analysis: The student writes a synthesis essay that integrates the findings from their human-sourced research, providing a nuanced analysis of the controversy.

4. Methods Appendix: The final paper includes a methods appendix that describes the AI prompts used and the guardrails or criteria employed to guide the AI's output.

4.6 Pattern 6: The Provenance Portfolio

A provenance portfolio shifts the entire assessment model for a course or term, making the process of creation as important as the final product. It is a powerful antidote to "contract cheating" and promotes a deep, metacognitive approach to learning.

Assignment Steps:

1. Collection Over Time: Throughout the term, students are required to collect and archive a wide range of materials related to their projects, including:

• Initial brainstorms and outlines.

• Successive drafts with version history.

• Feedback from instructors or peers.

• AI chat excerpts, including all prompts and outputs.

• A running list of sources and citations.

2. Final Reflection: At the end of the term, students submit their collected materials along with a final reflection. In this reflection, they analyze how GenAI affected their process, what they learned from using the tool, and what they would change for a future project. The reflection is a core component of the final grade.

V. Conclusion: Building a Sustainable Future for Learning

The mainstreaming of generative AI presents a defining challenge and a profound opportunity for higher education. This report has argued that the initial impulse to police and prohibit AI use is ultimately a losing strategy, one that fails to address the root causes of academic misconduct and may even undermine the value of a degree itself. The research and guidance from leading academic and quality assurance bodies consistently points toward a more proactive, pedagogical solution: one that moves beyond detection and focuses on the core principles of AI literacy, transparency, and process-oriented assessment.

By adopting the principles and assignment patterns outlined in this report, educators are not merely responding to a technological disruption; they are cultivating a new set of essential skills for their students. The ability to use, critique, and audit an intelligent system is a form of higher-order thinking that will be indispensable in a world increasingly shaped by algorithms. Furthermore, by embracing process-based and authentic assessments, educators can revitalize the human-centered elements of teaching and learning, ensuring that the classroom remains a space for genuine intellectual inquiry and growth. This is a continuous, evolving process, but the core principles laid out here provide a durable and sustainable foundation for a more purposeful and effective future for learning.

Appendix: Curated Resources for Educators

Key Organizations and Guides

• MLA-CCCC Task Force on AI & Writing: This joint task force from the Modern Language Association and the Conference on College Composition and Communication provides principles, policy scenarios, and a student guide for AI literacy.3

• TextGenEd (WAC Clearinghouse): An open-access, peer-reviewed collection of over 30 classroom assignments that integrate text-generation technologies, offering practical and creative ideas for the classroom.15

• Jisc / QAA Hubs: These hubs offer pragmatic design tools, guides for interactive oral assessments, and decision flowcharts for navigating assessment in an AI-integrated world.7

• EDUCAUSE Library: A comprehensive repository of resources, including reports, podcasts, and policy guides on AI use in higher education. Its resources cover a range of topics from AI ethics to practical policy frameworks.9

• Stanford HAI & Harvard Bok Center: These campus centers provide sample policies, assignment redesign guides, and a wealth of resources for faculty navigating the new landscape of AI in education.2

Sample AI Policy Templates

The following boilerplate language can be adapted for your course syllabus or assignment briefs to create a clear permission framework.

AI-Free Assignment Policy:

• "This assignment is designed to assess your unassisted mastery of foundational skills. The use of generative AI tools, including but not limited to ChatGPT, Claude, or similar services, is strictly forbidden at all stages of this assignment. Any use of these tools will be considered a violation of academic integrity."

AI-Assisted Assignment Policy:

• "You are permitted to use generative AI tools as a supplementary aid for this assignment. Permissible uses include brainstorming initial ideas, refining search queries, or checking for minor grammatical errors and typos. Using AI to generate full paragraphs, arguments, or the core content of the assignment is not permitted. You must include a brief statement in your submission, such as 'I used ChatGPT to brainstorm my initial topic ideas,' to disclose this use."

AI-Integrated Assignment Policy:

• "For this assignment, generative AI is a required tool. Your grade is not based on the AI’s output but on your ability to use the tool strategically and critically. You may use AI to generate multiple drafts or models, but you are responsible for critically evaluating, verifying, and significantly revising the output. You must attach an appendix or include a detailed AI use statement that explains what tool you used, how you used it, and how you verified its claims or revised its content."

Works cited

1. arXiv.org e-Print archive, accessed August 27, 2025, https://arxiv.org/

2. Stanford HAI: Home, accessed August 27, 2025, https://hai.stanford.edu/

3. MLA-CCCC Joint Task Force on Writing and AI, accessed August 27, 2025, https://aiandwriting.hcommons.org/

4. Jisc, accessed August 27, 2025, https://www.jisc.ac.uk/

5. The Quality Assurance Agency for Higher Education, accessed August 27, 2025, https://www.qaa.ac.uk/

6. EDUCAUSE Library | EDUCAUSE Library, accessed August 27, 2025, https://library.educause.edu/

7. Generative artificial intelligence - The Quality Assurance Agency for Higher Education, accessed August 27, 2025, https://www.qaa.ac.uk/sector-resources/generative-artificial-intelligence

8. Assuring higher education's quality in the age of AI is no easy task, accessed August 27, 2025, https://www.eua.eu/our-work/expert-voices/assuring-higher-educations-quality-in-the-age-of-ai-is-no-easy-task.html

9. Acceptable and Responsible Use Policies - EDUCAUSE Library, accessed August 27, 2025, https://library.educause.edu/topics/policy-and-law/acceptable-and-responsible-use-policies

10. The Derek Bok Center for Teaching and Learning - Harvard University, accessed August 27, 2025, https://bokcenter.harvard.edu/

11. UCLA Teaching & Learning Center: Homepage, accessed August 27, 2025, https://teaching.ucla.edu/

12. MLA-CCCC Task Force on Writing and AI - Todoele, accessed August 27, 2025, https://www.todoele.net/inteligencia-artificial-fuentes/mla-cccc-task-force-writing-and-ai

13. Student Guide to AI Literacy - MLA Style Center - Modern Language Association, accessed August 27, 2025, https://style.mla.org/student-guide-to-ai-literacy/

14. A Practical Guide to AI Literacy - EDUCAUSE Review, accessed August 27, 2025, https://er.educause.edu/podcasts/educause-shop-talk/2025/a-practical-guide-to-ai-literacy

15. Review of Annette Vee, Tim Laquintano, and Carly Schnitzler's TextGenEd: Teaching with Text Generation Technologies - Composition Forum, accessed August 27, 2025, https://compositionforum.com/issue/55/review-textgened/

16. Teaching@Sydney – Teaching@Sydney, accessed August 27, 2025, https://educational-innovation.sydney.edu.au/

17. Interactive Oral Assessment in practice – Teaching@Sydney, accessed August 27, 2025, https://educational-innovation.sydney.edu.au/teaching@sydney/interactive-oral-assessment-in-practice/

18. MLA-CCCC Joint Task Force on Writing and AI, accessed August 27, 2025, https://cccc.ncte.org/mla-cccc-joint-task-force-on-writing-and-ai

19. Continuing Experiments - The WAC Clearinghouse, accessed August 27, 2025, https://wac.colostate.edu/repository/collections/continuing-experiments/

20. Policy Publications | Stanford HAI - Stanford University, accessed August 27, 2025, https://hai.stanford.edu/policy/publications