Inspire vs Assure

Jason Blomquist; Liza Long; Joel Gladd

11 Inspire vs Assure

How This Chapter Fits in Part 2

The Need for Rethinking Assessment identified why finished products can no longer carry the whole assessment burden. Rethinking Bloom’s Taxonomy for the AI Era asked what thinking an assignment should make visible. This chapter turns those outcome decisions into a course map. The Inspire and Assure approach helps faculty consider how strongly an assessment motivates learning and how readily it can assure that learning occurred. The two-lane approach connects those decisions to the role and conditions of an assessment, including where learning is secured and where AI-supported practice can remain open.

The next chapter, AI Policy Frameworks and the AI Assessment Scale, uses that map to create assignment-level AI expectations.

Two Questions About Assessment

When faculty discuss AI and assessment, the conversation tends to center on security. The concerns are legitimate. Faculty want to prevent inappropriate AI use and catch violations. But security is only one design question. Before examining strategies for securing assessment, it helps to ask what will motivate students to do the work of learning and what will allow faculty to make a trustworthy judgment about that learning.

A 2025 global review from the Digital Education Council and Pearson maps 101 cases of AI-integrated assessment and identifies 14 design methodologies. Its dual-priority approach asks institutions to develop foundational human skills and AI fluency while protecting assessment validity.

Western Sydney University’s Inspire and Assure (IA) Approach offers a framework for holding those questions together. Cath Ellis introduced the model publicly with Jen Tindale and Brian Stout in 2025 as a refinement of the University of Sydney’s two-lane approach. The IA Approach asks faculty to examine two dimensions of an assessment: how strongly it motivates learning and how readily it can support a secure judgment that learning occurred.

The Inspire and Assure Approach

Inspire is the motivation axis. It runs from tasks that do little to motivate learning to tasks that make learning meaningful enough for students to engage, practice, explore, and grow.

Assure is the securability axis. It runs from tasks that are difficult to secure to tasks that make it easier for faculty to verify what students can actually do. Because the two axes intersect, a task can be high or low on either dimension. An assessment can inspire without providing strong assurance, assure without being especially motivating, do both, or do neither.

Western Sydney University’s Inspire and Assure Matrix
Quadrant	Motivation	Securability	Design Implication
Low Value	Lower	Harder to secure	The task is a strong candidate to stop or redesign.
Inspire	Higher	Harder to secure	The task motivates learning but provides limited assurance on its own.
Assure	Lower	Easier to secure	The task verifies learning but may need redesign to become more motivating or meaningful.
Inspire and Assure	Higher	Easier to secure	The task motivates learning and supports a trustworthy judgment about it.

The need for assurance has an empirical basis. Ellis and colleagues (2020) analyzed 221 orders from contract-cheating websites and 198 assessment tasks in which contract cheating had been detected. Tasks with none, some, or all five markers of authenticity were routinely outsourced. In these datasets, authenticity alone did not assure academic integrity. For this training, the implication is that meaningful design should be paired with selected verification moments. The study predates generative AI, but the underlying question remains: what evidence allows you to trust that learning occurred?

Assessments can contribute to both dimensions. The problem arises when every task carries the same security burden.

AI tools are also creating new possibilities for tasks high on the Inspire axis. A mixed-methods study by Sperber and colleagues (2025) found that students often valued AI and peer feedback together. Their reflections showed critical evaluation of AI feedback, emerging AI literacy, and writerly agency. When the purpose is engagement and growth, AI can be part of the learning process itself.

The Two-Lane Approach

The IA Approach and the two-lane approach overlap, but they ask different design questions. IA examines motivation and securability. The two-lane approach connects the role of an assessment to the conditions under which students complete it.

The University of Sydney describes two broad lanes:

Lane 1: Secure. Students complete the assessment under controlled conditions. The purpose is to form a trustworthy judgment about what students can do. Lane 1 includes exams, in-class writing, observed performances, oral defenses, lab practicals, clinical demonstrations, and other moments where independent capability needs to be verified.

Lane 2: Open. Students complete assessment for and as learning in a more open environment, often with access to AI and other tools. The purpose is learning, practice, development, and authentic participation in a discipline where AI tools are increasingly common. Lane 2 includes drafts, projects, peer review, AI-supported brainstorming, reflective work, simulations, and process-oriented tasks.

For practical course mapping, this training uses the crosswalk below as a local design aid. It should not be mistaken for part of either source framework.

Practical Crosswalk: Faculty Questions
IA Pattern	Faculty Question
High inspiration, harder to secure	How does this help students practice, explore, or grow, and what additional evidence would assure attainment?
Easier to secure	How will I know the student can actually do the thing, and how might the task become more motivating?
High inspiration and assurance	Can open preparation and a secured checkpoint work together?

Practical Crosswalk: Conditions and Examples
IA Pattern	Common Student-Facing Condition	Example
High inspiration, harder to secure	Lane 2: open or AI-supported	Brainstorming, drafts, practice problems, AI feedback, peer review
Easier to secure	Lane 1: controlled or secured	In-class writing, lab practical, oral defense, live presentation
High inspiration and assurance	Lane 2 work plus a Lane 1 checkpoint	AI-supported project with a short live defense

The table shows a common pattern, and other combinations are possible. A Lane 1 assessment can include AI if the learning outcome being verified includes responsible AI use. Sydney’s detailed two-lane guidance describes controlled AI use as one option for secured assessment. An interactive oral or mock-client exchange can then test the student’s judgment while the instructor observes the work and asks follow-up questions.

Lane 2 also needs structure. Curtin University’s Assessment 2030 materials frame Lane 2 as assessment for learning, where students develop professional capabilities in environments that may include digital and AI tools. Curtin’s 2026 guidance emphasizes process, verification, evaluative judgment, and student accountability. Students may have access to AI, but they still need to defend choices, explain or reflect on their process when appropriate, and meet the learning outcomes.

The two-lane approach gives students a simple map. Some tasks are secured because the course needs trustworthy evidence of learning. Other tasks are open because the course is helping them practice, experiment, revise, and learn with the tools available in the world they are entering.

Compliance vs. Intrinsic Motivation

This training applies the Inspire axis to a related question: how do we encourage the behaviors we want from students?

Compliance-based approaches rely on rules and monitoring. We tell students what they cannot do, watch for violations, and impose sanctions.

Motivation-based approaches design learning experiences that students find genuinely valuable. Consider the difference between a student designing a marketing plan for a local business she cares about and a student writing a generic five-paragraph essay on a topic she did not choose. Research on autonomy-supportive teaching in higher education finds positive associations with student motivation and engagement, although the association with academic performance is smaller.

Marc Watkins argues that adversarial responses to AI undermine teaching relationships. When every assignment is treated as a potential integrity violation, the classroom can become a site of suspicion that competes with teaching.

Effective course design combines motivation and accountability. Motivation-based approaches can support many learning activities. Accountability structures are reserved for the moments where evidence of individual mastery matters.

Faculty Spotlight: Marcus Hickens, Business

Marcus teaches Principles of Management and has been cautious about AI from the start. His course includes written reflections on readings, weekly discussion posts, practice management scenarios, a group consulting project, and a midterm case presentation. When he first encountered the IA Approach, he realized he had been treating every assessment as if it carried the same security burden. Every task had AI restrictions. Every task required an originality statement. He was spending hours scanning discussion posts for AI-generated language.

The discussion posts stood out. Students respond to a management scenario each week and react to each other’s interpretations. Marcus placed them high on the Inspire axis but found them difficult to secure. He kept them in Lane 2. Students could use AI to brainstorm or refine a point if they wanted, as long as the conversation stayed substantive.

The midterm case presentation went the other direction. Students present a consulting recommendation to a panel and defend their reasoning in real time. The task is easier to secure and can still motivate students through a realistic professional exchange, placing it in the Inspire and Assure quadrant. It is also a Lane 1 moment. AI can help students prepare, but the live defense gives Marcus evidence of whether they understand the recommendation.

The group consulting project needed a mixed design. Its real client makes it highly motivating, but the open project process is difficult to secure. Marcus keeps the process in Lane 2 so students can research, brainstorm, draft, and revise with tools. He then adds a Lane 1 checkpoint: each student gives a short oral explanation of one decision they personally contributed to the project. Together, the open project and secured checkpoint move the design toward the Inspire and Assure quadrant.

To test his thinking, Marcus pasted his assessment list into an AI chatbot and asked it to rate each task for motivation and securability, place it in the IA matrix, and then suggest a lane. The AI classified the group consulting project as an Inspire task in Lane 2 because collaboration and process were central. Marcus partly disagreed. The open process fit Lane 2, but the final deliverable had real stakes and the course still needed evidence of each student’s judgment. He kept the project open and retained the Lane 1 checkpoint. The exercise was still useful because it gave him a starting point for judgment.

Options for Accountability and Assurance

When an assessment needs to assure learning, the next question is how to build in accountability. Controlled or supervised strategies can create a Lane 1 condition. Process documentation, collaboration, and version history can provide supplementary evidence, but they do not secure an assessment on their own. The right approach depends on your teaching context, the format of the assessment, and what you can realistically sustain across a full semester.

The table below compares four common strategies. Each involves a core tradeoff. Each fits a different course pattern.

Accountability and Assurance Strategies: When to Use
Strategy	When to Use
Synchronous/Proctored	High-stakes exams, oral defenses, classroom debates, certification checkpoints
Process Documentation	Papers, projects, and portfolios where the development process matters
Collaborative Tasks	Group projects, peer review, discussion-based assessments
Version History/Labor Tracking	Written assignments in cloud-based tools (Google Docs, Word 365)

Accountability and Assurance Strategies: Tradeoffs
Strategy	Key Tradeoff
Synchronous/Proctored	Security vs. accessibility and student wellbeing
Process Documentation	Accountability vs. grading burden
Collaborative Tasks	Peer accountability vs. uneven group dynamics
Version History/Labor Tracking	Transparency vs. perceived surveillance

Accountability and Assurance Strategies: Risks to Watch
Strategy	What to Watch For
Synchronous/Proctored	Test anxiety, scheduling barriers, logistical cost at scale
Process Documentation	Requires clear scaffolding, increases workload for students and faculty
Collaborative Tasks	Coordination challenges online; stronger when paired with individual explanation or defense
Version History/Labor Tracking	Can feel invasive, can be gamed, adds technical complexity

The choice among these strategies depends on context. If you teach fully online, synchronous proctoring may be impractical, but process documentation can work well for written assignments. If you teach large lecture sections, version tracking can create an unsustainable review load. A single well-designed oral defense at the end of the term might be more manageable.

Oral and live formats can provide another source of evidence under secured conditions. Classroom debates, viva-style defenses, and live presentations allow faculty to ask responsive questions and hear how students explain or apply their thinking beyond a written submission. A systematic review of oral assessment also emphasizes the importance of structured practice, feedback, and clear expectations. These formats can be paired with written assignments, but faculty still need to consider alignment, accommodations, anxiety, and unfamiliarity. The writing can become the preparation, and the live component can become the verification.

Lab-based courses can use collaborative tasks with live demonstration components to provide accountability without extra infrastructure. Anna Mills has documented what a layered approach looks like in practice, combining multiple lighter-touch strategies such as peer review, social annotation, writing process assignments, and student choice so that a course does not rely on one tool for accountability.

Faculty can combine two of these strategies across a semester. A writing-intensive course might pair process documentation on the major papers with a brief synchronous presentation at the end. A quantitative course might rely on proctored exams for the midterm and final while leaving all practice sets in Lane 2.

The goal is to match the level of security to the purpose of the assessment and to be honest about what you can sustain.

Ethical Tradeoffs

Every strategy for accountability and assurance involves tradeoffs. No approach is both perfectly secure and completely unobtrusive.

Equity implications. Does a given strategy disadvantage certain students? Synchronous assessment and remote proctoring do not carry identical risks. A 2024 scoping review of remote proctoring found recurring student concerns involving privacy, technical problems, fairness, and stress. Any secured format should be checked for accommodation and scheduling barriers. Access to AI also varies. Curtin’s 2026 guidance (PDF) cautions faculty against grading outputs that students can achieve only with premium tools.

Trust dynamics. What message does a strategy send about your relationship with students? Excessive security measures can communicate distrust. Marc Watkins argues that surveillance-based approaches, including some forms of process tracking, can undermine the trust that productive teaching relationships depend on. Treat this as an ethical design question rather than a causal conclusion. Later chapters will explore process tracking in more depth.

Proportionality. Is the level of security appropriate to the stakes? A low-stakes practice assignment probably does not need the same security measures as a certification exam. Matching security to stakes shows students you are being deliberate about the distinction.

Sustainability. Can you maintain this approach across multiple sections and semesters? Faculty workload and scalability should be part of the choice from the beginning.

Bidirectional transparency. The ethical conversation about AI and assessment is reciprocal. If we expect students to be transparent about their AI use, faculty should consider their own practices too. Are you using AI to generate quiz questions? To provide feedback? To grade? These are legitimate choices, but they carry implications for the trust you are asking students to extend.

Focusing Your Accountability Energy

One practical implication of the IA Approach is that the framework does not ask you to secure every activity. It gives you permission to let go of low-value tasks and focus energy on the assessments that matter most.

Ellis and Jason Lodge put the core idea simply in a widely shared 2024 LinkedIn post. Stop looking for evidence of cheating. Start looking for evidence of learning. That shift in attention changes how you spend your time and what your course feels like to students.

In practice, activities high on the Inspire axis may occupy much of the semester. Students write discussion posts, complete practice sets, draft and revise, and collaborate with peers. Those are often Lane 2 activities. A few strategic assurance checkpoints, placed at key points in the course, provide evidence that learning has occurred. Those assessments should receive the most deliberate security measures. In this design, open practice provides preparation for secured moments. The weekly management discussions in Marcus’s course, for example, build the analytical habits students need for the midterm case presentation.

This approach reserves security measures for selected assessments and focuses faculty energy where it counts.

Try This: Map One Assessment

Step 1. Choose one assessment from one of your courses. Briefly describe what students do and what the assessment measures.

Step 2. Rate the assessment on the two IA dimensions:

Motivation: How strongly does the task motivate students to do the work of learning?
Securability: How easy or difficult is it to make a trustworthy judgment about what the student can do?

Use those two judgments to place the task in one quadrant: Low Value, Inspire, Assure, or Inspire and Assure.

Step 3. Classify the assessment by lane:

Lane 1: students complete the task, or one part of the task, in a controlled or secured setting.
Lane 2: students complete the task in an open setting where AI and other tools may be available.
A design combining both: students work openly for one part of the task and complete a secured explanation, demonstration, or defense for another.

Here is one completed sample row from Marcus’s course.

Assessment: Weekly management discussion post
IA Judgment: Inspire quadrant: high motivation, harder to secure
Lane: Lane 2: open practice
Rationale and Tradeoff: The discussion supports application and peer exchange. Proportionality matters more than securing every weekly response.

Step 4. Paste the assessment description into an AI chatbot with this prompt:

Use the R-T-C-I pattern from Part 1 when you do this. Give the AI the role of assessment-design reviewer, define the task, provide course context, and ask for a concise table or list you can evaluate.

I am reviewing one course assessment with Western Sydney University’s Inspire and Assure matrix and the University of Sydney’s two-lane approach. First, rate the task’s motivation from low to high and its securability from hard to easy. Place it in the Low Value, Inspire, Assure, or Inspire and Assure quadrant. Then suggest whether it belongs in Lane 1, Lane 2, or a design combining components from both lanes. Explain your reasoning and identify one assumption I should question.

Step 5. Compare the AI’s categorization with your own judgment. Consider one ethical tradeoff involving equity, trust, proportionality, or sustainability. Revise your classification or rationale if the tradeoff changes your decision.

Looking Ahead

This chapter helped you examine motivation and securability, then map one assessment’s role and completion conditions. The next chapter, AI Policy Frameworks and the AI Assessment Scale, moves to the assignment level. It introduces the AI Assessment Scale, a framework for specifying exactly how AI may be used on a task.

References

Curtin University. (n.d.). Assessment 2030. Source link

Curtin University. (2026). Assessment 2030: GenAI guidance for Curtin educators in 2026. PDF source

Digital Education Council. (2025, July 7). The next era of assessment: A global review of AI in assessment design. In partnership with Pearson. Source link

Ellis, C., van Haeringen, K., Harper, R., Bretag, T., Zucker, I., McBride, S., Rozenberg, P., Newton, P., & Saddiqui, S. (2020). Does authentic assessment assure academic integrity? Evidence from contract cheating data. Higher Education Research & Development, 39(3), 454-469. DOI record

Ellis, C. (with J. Tindale & B. Stout). (2025, December 1). Reimagining assessment for a GenAI world: Introducing Western Sydney University’s IA approach. LinkedIn. Source link

Ellis, C. (with J. M. Lodge). (2024, July 8). Stop looking for evidence of cheating with AI and start looking for evidence of learning. LinkedIn. Source link

Long, L. (2025, May 22). ChatGPT: The teacher ethics edition. Artisanal Intelligence. Source link

Marano, E., Newton, P. M., Birch, Z., Croombs, M., Gilbert, C., & Draper, M. J. (2024). What is the student experience of remote proctoring? A pragmatic scoping review. Higher Education Quarterly, 78(3), 1031-1047. DOI record

Mills, A. (2026, February 7). A brief update: Why I’m still using AI detection after all, alongside many other strategies. Anna Mills’ Substack. Source link

Okada, R. (2023). Effects of perceived autonomy support on academic achievement and motivation among higher education students: A meta-analysis. Japanese Psychological Research, 65(3), 230-242. DOI record

Sperber, L., MacArthur, M., Minnillo, S., Stillman, N., & Whithaus, C. (2025). Peer and AI review + reflection (PAIRR): A human-centered approach to formative assessment. Computers and Composition, 76, 102921. DOI record

Stephenson, Z., Johnson-Glauch, N., & Cruchley, S. (2025). Interventions and facilitators of oral assessment performance in higher education: A systematic review. Assessment & Evaluation in Higher Education, 50(7), 1140-1153. DOI record

University of Sydney. (2024, November 27). University of Sydney’s AI assessment policy: Protecting integrity and empowering students. Source link

University of Sydney. (2024). Frequently asked questions about the two-lane approach to assessment in the age of AI [Archived page]. Internet Archive. Source link

Watkins, M. (2025, December 1). Our response to AI cannot be adversarial. Rhetorica. Source link

Watkins, M. (2025, March 16). When algorithms watch you write. Rhetorica. Source link

License

Icon for the Creative Commons Attribution-NonCommercial 4.0 International License

A Guide to Teaching and Learning with Artificial Intelligence Copyright © by Jason Blomquist; Liza Long; and Joel Gladd is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License, except where otherwise noted.