AI and assessment

Designing assessment for achievement and demonstration of learning outcomes

A key purpose of assessment is to yield evidence of student learning. This evidence depends on submitted work accurately representing students’ efforts and human abilities and/or knowledge.

Therefore use of generative AI technologies to produce text and other media as part of student submissions (or, indeed, as part of the process of developing such submissions) needs to be thoughtfully supported to ensure responsible use and clear demonstration of human achievements.

The proliferation of AI in education provides valuable opportunities to consider why we are assessing our students, what is being evaluated, and how evidence of learning is being gathered. Providing a clear rationale for assessment design and associated integration of AI will help students to understand why they are undertaking such activities and to see the value in responsible uses of AI technologies.

Chief Examiners, who are responsible for the assessment regimes in units, must specify where and how AI may be used in assessments. AI use also needs to be clearly and openly acknowledged by students, in line with what has been specified by the Chief Examiner. AI use should align with the Monash University institutional position of ethical and responsible AI use and the Academic Integrity Policy. Any restrictions placed on AI use must be for clearly articulated pedagogical or accreditation reasons and accompanied by appropriate assessment security measures.

Considerations for assessment design

In considering the role of AI technologies in assessment, a key starting point is to determine if you will be:

Collecting evidence to confirm unassisted human abilities or to confirm knowledge and skills that can be collaboratively achieved (with other humans and/or other intelligences).
Integrating AI collaboration into the assessment by designing assessment tasks that require and encourage appropriate use of AI.
Designing assessment tasks that make the use of AI less relevant, such as by focusing on embedded, continuous assessment, highly contextualised and personal experience, or explicit demonstration of human capabilities.
Designing assessment tasks that demonstrate individual and independent human knowledge, or that bracket AI out with appropriate measures in place.

Test current assessment design View

Key Question: Given student access to generative AI (GenAI), does my assessment still provide valid evidence of learning?

It can be useful to ‘test’ whether your assessments are designed in a way where they can be easily completed by GenAI tools. Try the following protocol to test your assessment design with current generative AI tools:

Paste the assignment prompt into a GenAI tool
- Use the Monash enterprise tools such as CoPilot or Google Gemini, or free versions of ChatGPT or Claude (recognising the input may be contributing to training the model).
- Observe the initial output without additional context.
- Ask: What grade would this receive using your current rubric?
Simulate a student scenario
- Assign the AI a role (e.g., “You are a 3rd-year student majoring in history.”).
- Tell the AI that you will be asking it to perform a task and that you will provide it the reference material in steps. (e.g., “I will provide the assignment guidance document, a rubric against which it will be marked, and a reference case study in separate prompts. Once you have read them, please write a submission for the assignment.”)
  - Provide the AI with all the information a student would have:
  - Detailed instructions and guides
  - Marking rubric
  - Reference materials or case studies
- Break the task into subtasks if applicable.
Evaluate the AI’s submission
- Review the structure, accuracy, and depth.
- Provide feedback to the AI and observe its revisions.
- Ask: Does the AI improve with feedback?
- Ask: Is the process replicable by students?
Assess the risk level
- Determine if the AI produces high-quality work with minimal input.
- Ask: Is there evidence of how a student arrived at the final product?
- Ask: Is evidence of unassisted human abilities required for all of this task?

Consider the results

If the GenAI performed well with the assignment, you may need to consider updating your assessment design. Even if the GenAI did not perform well, it does not necessarily mean that your assessment is “safe.” AI capabilities are changing and improving rapidly.

You should periodically review the learning outcomes, task and marking procedures to ensure your assessment design remains valid and contemporary.

For many assignments, Large Language Models (LLM) such as Google Gemini, Microsoft Copilot or ChatGPT will produce a reasonable-looking structure but will often lack precise details and may contain factual inaccuracies or non-existent references.
Generally, GenAI tools are able to follow simple instructions and provide answers that have a layer of correctness but may not be strong in contextualising responses.
- LLMs generate responses by predicting statistically likely continuations based on training patterns, rather than through deliberate reasoning like humans.
- While they can follow instructions effectively, they may struggle to prioritise competing or conflicting instructions without explicit human guidance.
- There is a “dice roll” associated with each AI interaction. The same prompt will provide a different response each time; successive prompts multiply potential permutations.
- Different AI tools have different capabilities and better versions may be behind paywalls.

Confirm and coordinate the learning outcomes View

Key Question: Given student access to generative AI (Gen AI) to assist with aspects of the task, does my assessment in its current form still provide valid evidence of learning?

Review learning outcomes to ensure they describe what a student will be able to do in a demonstrable and measurable way.

Consider if the current learning outcomes are current, sustainable and future oriented.
Consider where you require students to learn about and demonstrate the responsible use of AI technologies.

Coordinate where foundational knowledge and skills are taught in the course to minimise duplication and identify gaps in the student learning journey. Then consider where and how those foundations are developed into more complex knowledge and skills across the curriculum including when and where contemporary tools and approaches will be incorporated.

Open the accordions for more detail.

It may now be necessary to target forms of knowledge and expression that are more difficult for generative AI technologies to easily accomplish, such as critical thinking, evaluation or creativity.
Select the plus icons for a definition of each term

Integrate AI into learning outcomes

Consider where AI technologies fit in your course and unit learning outcomes. Knowledge and skills relating to emerging technologies can be integrated in either an explicit or implicit way. Consider:

Does the discipline require a specific learning outcome related to emerging technologies?
How sustainable or future oriented are the learning outcomes?

Examples of explicit and implicit framing.

Explicit example 1: Demonstrate broad knowledge and technical skills in [discipline], including the use of digital technologies and techniques, to provide solutions relevant to diverse stakeholders in dynamic environments

Explicit example 2: Apply relevant knowledge, skills, critical thinking and curiosity to effectively address discipline challenges using current and emerging tools and adapting solutions to be context-sensitive and suitable for the intended audience.

Implicit example 1: Collaborate with others to create meaningful solutions to [discipline]] challenges.

Implicit example 2: Demonstrate advanced professionalism through a commitment to lifelong learning, ethical practice and advocacy for the [discipline].

Next, consider whether the course or unit learning outcomes should be achieved individually or collectively. Is there scope and space for outcomes to be achieved:

in collaboration with other students
with AI assistance or
to intentionally exclude AI assistance and peers to demonstrate individual human skill?

Examples of individual and team learning outcome framing.

Individual example: Communicate effectively and efficiently.

Ambiguous example: Conduct research and be skilled at accessing, critically appraising and applying the best available evidence to practice.

Team example: Work collaboratively in a range of settings.

As with all learning outcomes you should map and coordinate where foundational knowledge and skills are taught in the course to minimise duplication and identify gaps. Then consider where and how those foundations can be developed into more complex knowledge and skills across the curriculum.

Refer to Learning outcomes - integrate AI into learning outcomes for a discussion between Professor Ari Seligmann and Associate Professor Tim Fawns about adapting learning outcomes at both a unit and course level. As well as more guidance about designing learning outcomes that align with taxonomies to describe the desired performance or behaviours that you want to observe as evidence of learning.

Review the task View

Key question: Is the product submitted for marking easily completed by GenAI alone?

While targeting higher order thinking may reduce the risk of misuse of generative AI tools, it is not sufficient to only consider the kind of thinking required to produce a response to an assessment task (e.g. by targeting higher order categories of Bloom's taxonomy). For example, in some circumstances, it is possible for Gen AI to represent or imitate each of these categories of learning (e.g. it can create new ideas by synthesising unusual combinations of information sources).

However, there are other assessment characteristics that can be associated with higher and lower risk of inappropriate use of generative AI. The following diagrams illustrate some of these characteristics in relation to risk.

Consider tasks that make the use of AI less relevant, such as by focusing on embedded, continuous assessment, highly contextualised and personal experience, or explicit demonstration of human capabilities.
Consider tasks that highlight the processes students used to produce their assessment products such as portfolios, journals, video logs, tracked changes. Have students indicate which aspects of the product were created by them, AI, or in collaboration.
Consider tasks that leverage the collaborative nature of student/AI interactions, e.g., co-creating a document or program, critiquing AI output, generating alternative solutions, or addressing the weakness of AI.

Open the accordions for more detail.

Consider alternative assessment formats

Some formats are more difficult for AI tools to generate than others, with text and code currently the simplest.

The table below shows example assessments that are currently more or less difficult for Copilot and GPT-5 (many of these examples are also applicable to other generative AI tools).

Difficulty for AI technologies	Task	Suggestion
	Closed answers e.g. MCQ or short answer exam questions that ask students to define, list or reproduce.	LLMs excel at factual recall and synthesising information to produce convincing answers to closed or abstract questions. Where feasible, consider alternative formats that align with unit learning outcomes.
	Essays without personalised application	LLMs can produce convincing text in many different genres and styles as requested (although it may find accurate referencing difficult). Modify questions to require personalised application. Combine with modifications such as assessing process or personal reflection.
	Essays with personalised application e.g. local case studies, unique data or class discussions.	LLMs are poor at genuine personalised application or contextualisation of ideas. However, caution is warranted as surface-level generic responses that include plausible but fabricated details can be easily generated, Students may also find personalised writing difficult and scaffolding this kind of writing is important. Combine with modifications such as assessing process or personal reflection and scaffold students’ abilities to write in this way.
	Showing individual working process e.g. research logs, iterative drafts with feedback, step-by-step problem solving.	LLMs do not reveal how outputs are generated, so it cannot produce an account of its own iterative process. Replicating that would be significant work for students. Consider introducing staged submissions where students explicitly respond to feedback on drafts, or where they are asked to submit a portfolio of the work they have done on the way to their final submitted work.
	Showing collaborative working process	Producing a misleading account of how students worked together to produce an output would involve significant work. Consider group tasks (with or without a final, individual submission) where students are asked to reflect on the process of completing the work and what they have learned from it.
	Multimodal artefacts (e.g. a document that contains text + images) as creative response.	While LLMs can generate text and images in many cases thoughtful engagement with the material is required to compose something meaningful that is more than the sum of its parts. While it may easily generate code or captions for images, producing a coherent multimodal artifact is currently outside its domain. Consider asking students to produce images (e.g. an infographic) or video recordings as part or all of a response to a question. These should not just be reproducing text in a different format (e.g. a recording of reading text aloud) but should make use of the format as an alternative way of expressing ideas. Note: Multimodal work is often more ambiguous than text and may require more marker training and time. It may also require more technical support.
	Interview / viva voce	Currently not possible for AI technologies to replicate in real time. Consider asking students to participate in a live (synchronous) conversation (on campus or online) about what they have learned or about their submitted work. Note: This may require scheduling for each student, and may involve a greater time commitment for marking.

Select the plus icons for a definition of each term

Review the marking procedures View

Key question: Are the marking criteria and weighting distribution valid evidence of learning?

Marking represents judgement of quality based on expressed criteria and provides indicators of achievement of learning outcomes.

Consider whether the marking criteria relate to core skills where human competency is required, rather than focusing on easily automated tasks. As Gen AI technologies become increasingly embedded in tools and services, traditional criteria such as spelling, grammar, and basic writing elements, may no longer serve as meaningful indicators of student learning.

Marking criteria:

should prioritise higher order thinking skills that demonstrate critical thinking, analysis, and creative problem-solving capabilities that require human intelligence.
could shift focus from evaluating the quality output to assessing the process: students' abilities to effectively prompt, critique, edit, and meaningfully integrate AI contributions into their work.

Gen AI tools can potentially generate unexpected high-quality outcomes that don't fit traditional marking schemes. To address this complexity, rubrics could be structured to evaluate content knowledge, learning processes, and the strategic use of AI tools.

Open the accordions for more detail.

Consider whether you are ignoring interactions between students and tools. Are your attributing evidence of learning outcomes to individuals when you should be attributing them to a system or collaboration?
Part of a valid assessment argument is being able to determine who produced the work product. Do your task conditions make it possible to determine who produced what?
- Ensure students are aware of the task conditions relating to the use of AI, and that these are clearly stated in Moodle.
- If evidence of individual and independent human knowledge or skills is required, then consider how to ensure visibility of learning.
Marking processes can be quite different from marking products. Some key things to pay attention to are: attribution (who did what), sequence (when did it happen and in what order), and interaction (how are actions responding to and building upon one another).
To make process marking viable, tasks can be structured with sequential steps or milestone checkpoints that naturally document student thinking and problem-solving approaches, such as reflection logs, decision trees, or staged submissions.
In many higher education models, high stakes, heavily weighted summative assessment is the primary source of data about student performance and progress. Where the evidence used as a basis for judging student performance has been partly or wholly generated by technologies, this can lead to misrepresentation of student capability. Adopting a varied approach in designing an assessment regime, where data about student performance is gathered from formative and diagnostic tasks, can help to create a strong relationship with students, motivate them in their learning and provide information about whether summative assessment performances are consistent with other work.
While generative AI tools pose a significant challenge for some assessment formats when used for high stakes, summative assessment of learning (e.g. end of unit essays, etc.), they may be less problematic for alternative formats and lower stakes, formative assessment for learning.
Selecting a range of tasks with varied weighting can allow students to demonstrate learning in a lower-risk environment, giving you the opportunity to gauge your students' understanding while also lowering the incentive for them to rely on tools to generate responses.
Formative assessment that isn’t weighted also allows for opportunities to check understanding and build a picture of student capability. Where formative assessment helps students prepare for summative assessment, it may also increase their confidence to engage authentically in those summative assessment tasks.

Contributions

This content was produced with contributions from the AI in Education Learning circle: Natalia Antolak-Saper, Kiri Beilby, Brendan Boniface, Dana Bui, Paul Burgess, Aamir Cheema, Michael Crocco, Robbie Fordyce, Kirstie Galbraith, Gaye Lansdell, Caryn Lim, Joel Moore, Amelia Nathania, Sadia Nawaz, Limalini Raveendran, Tridib Saha, Carmen Sapsed, Brendan Shannon, Kimberly Soh, Zach Swiecki, Thao Vu, Peter Wagstaff, Estelle Wallingford, Pauline Wong, Farid Zaid.

Considerations for assessment conditions

The following section provides guidance on the assessment conditions including measures for visible evidence of learning.

AI and Generative AI tools may be used within educator's guidelines View

Evidence of unassisted human knowledge and skill is required View

There are many possible reasons for excluding AI from assessment, including external accreditation and other pedagogical considerations. Make sure you have properly identified why you want to keep AI out and communicate this to students as part of the assessment conditions.

Consider when do students need to demonstrate unassisted individual knowledge and skills?
What assessment design can provide visible evidence of unassisted learning and skills?

The conditions for restriction are to be clearly communicated on Moodle or within the assessment instructions. See how to use AI Statements as the primary way that educators communicate the requirements for AI usage in accordance with current Monash procedures.

Where it has been determined that the use of AI is not permitted in an assessment aimed at demonstrating individual human capacities, then educators may want to use mechanisms to help ensure the protection of the assessment from AI use. However, as AI is increasingly integrated into a wide range of software and tools, bracketing AI out of assessments may be challenging.

Since 2023 there have been a variety of responses to AI and some institutions responded to AI technologies by prohibiting their use, blocking access to them from university networks and reverting to traditional modes of assessment such as in-person pen and paper exams or oral exams.
High stakes exams do not promote authentic learning experiences for students and limit opportunities for assessment via multiple modalities or for feedback that can contribute meaningfully to learning.
Accessibility and inclusivity principles are also compromised in high stakes exams. For example, hand-written and oral exams are known to significantly disadvantage students with physical or mental disabilities (see for example Nienminen 2021 or 2022 ).
Attempts to block remote access to technology can often easily be bypassed by students. Attempting to use other technology to detect instances of this behaviour is unlikely to be effective or sustainable.
For ideas about effective alternatives see more on Active learning on Teach HQ.

Visible evidence of learning and contemporary academic integrity View

Considerations for visible evidence of learning (as an alternative framing to assessment security) and contemporary academic integrity.

Academic Integrity relates to the ethical, responsible maintenance of academic conventions and practices. This is distinct from assessment security which relates to protecting tasks from undesirable actions.

"Academic integrity focuses on equipping learners with the capabilities and values necessary to conduct ethical scholarship. In contrast, assessment security focuses on hardening assessment against attempts to cheat, and on detecting any cheating that has occurred" (Dawson et al.).

Distinguishing academic integrity from assessment security allows us to separate out the educative response to building good practices from the locking down of assessment. However, at Monash, we seek to move away from ideas of security, enforcement, policing, and the detection of negative or undesirable behaviours, to consider ways in which we can detect appropriate learning. As commentators such as Dawson have argued, this is more aligned with the concept of assessment validity.

Working to make evidence of learning more visible plays a crucial role in supporting academic integrity by making the learning process more transparent and observable, enabling educators to assess not just the final product but the development of understanding over time. This helps affirm the authenticity of student work and, where accompanied by appropriate scaffolding and conversation, reinforces ethical learning behaviors.

The following strategies help enhance the visibility of student learning.

Assessment is conducted in real-time with the educator physically or virtually observing the participant’s performance, ensuring authenticity of the work.
Participants articulate their understanding verbally, making it difficult to rely on AI-generated content and allowing educators to assess depth of knowledge through a dialogue and questioning about the topic.
Participants submit drafts, notes, or reflections that show the evolution of their thinking, helping verify the originality and authorship of the work.
Work is collected at multiple points across a semester to track progress, making sudden, unexplained jumps in quality more noticeable.
Participants engage in collaborative tasks or receive iterative feedback, embedding educator oversight into the learning and assessment process.
Physical or digital restrictions (e.g., supervised settings, lockdown browsers) are used to limit access to unauthorized tools during assessments.
Tasks are completed using non-digital methods (e.g., handwritten essays, in-person performances), reducing the opportunity for AI assistance.

Assessment conditions

Chief Examiners have overarching responsibility for designing and setting assessment conditions, including communicating and verifying the responsible use of AI within assessment tasks.

The conditions for AI use, or restriction, are to be clearly communicated on Moodle or within the assessment instructions. See how to use AI Statements as the primary ways that educators communicate the requirements for AI usage in accordance with current Monash procedures.

As outlined in the Student Academic Integrity Procedure, students must not breach assessment conditions set by the Chief Examiner. Any work submitted for a mark must:

represent a sincere demonstration of your human efforts, skills and subject knowledge that you will be accountable for.
adhere to the guidelines for AI use set for the assessment task.
reflect the University’s commitment to academic integrity and ethical behaviour.

Breach of assessment conditions may be considered a breach of academic integrity. Where AI use is permitted, students must acknowledge the use of AI technologies in the creation of work.

For full guidance on Monash processes and systems access academic integrity (Monash staff only).

What constitutes a breach of assessment conditions?

According to the Student Academic Integrity Procedure Procedure: “Students must not breach the conditions for assessments set by chief examiners in the learning management system or within the assessment instructions.”

Conditions may include but are not limited to:

"only accessing authorised materials or not accessing any materials during an assessment (e.g. specifically permitted items
or a closed-book quiz or test);
only using specifically authorised tools or devices (e.g. using only approved calculators);
completing an assessment as a group or individually;
not using generative artificial intelligence tools; or
acknowledging the use of generative artificial intelligence tools in the creation of work."

In other words, suspicions and investigations of breach of assessment conditions involving AI use may be triggered by one of two key conditions.

1. When a student uses generative artificial intelligence tools when the Chief Examiner has specified in the assessment documentation in the learning management system that it cannot be used
Brief explanation	Using AI when specifically instructed not to (this includes misconduct when false declarations are made about how AI has been used ) or in ways that have been specifically restricted.
Colloquial explanation	“Did something that you were specifically told not to” or “Used AI when you were not supposed to.”
Ways to avoid breaches	Educators carefully explaining the conditions and expectations of the assessment. Students carefully following the instructions provided for the assessment task and asking questions if unsure.

1. When a student uses generative artificial intelligence tools when the Chief Examiner has specified in the assessment documentation in the learning management system that it cannot be used

Brief explanation

Using AI when specifically instructed not to (this includes misconduct when false declarations are made about how AI has been used ) or in ways that have been specifically restricted.

Colloquial explanation

“Did something that you were specifically told not to” or “Used AI when you were not supposed to.”

Ways to avoid breaches

Educators carefully explaining the conditions and expectations of the assessment.

Students carefully following the instructions provided for the assessment task and asking questions if unsure.

2. When a student fails to appropriately acknowledge the use of generative artificial intelligence tools in the creation of work submitted for assessment.
Brief explanation	Not appropriately acknowledging use of AI in the production of an assessment leading people to believe that you did all of the work yourself without assistance and that the submission represents your individual human abilities.
Colloquial explanation	“Got AI assistance to complete the task(s) but did not credit your AI collaborators and made it seem like you did it all yourself.”
Ways to avoid breaches	Educators carefully explaining the conditions of AI use. Students carefully documenting the assessment production process and truthfully explaining the processes and AI tools used. This should be done in the acknowledgement section of assessments, where included.

2. When a student fails to appropriately acknowledge the use of generative artificial intelligence tools in the creation of work submitted for assessment.

Brief explanation

Not appropriately acknowledging use of AI in the production of an assessment leading people to believe that you did all of the work yourself without assistance and that the submission represents your individual human abilities.

Colloquial explanation

“Got AI assistance to complete the task(s) but did not credit your AI collaborators and made it seem like you did it all yourself.”

Ways to avoid breaches

Educators carefully explaining the conditions of AI use.

Students carefully documenting the assessment production process and truthfully explaining the processes and AI tools used. This should be done in the acknowledgement section of assessments, where included.

Monash is committed to responsible and ethical use of AI that is transparent, explainable, documented and acknowledged. Moreover, humans who submit assessment items will be held fully responsible for the accuracy, content and academic integrity of work submitted.

Where it has been determined that the use of AI is not permitted in an assessment aimed at demonstrating unassisted human capacities, then educators may want to use mechanisms to help ensure the security of the assessment from AI use. However, as AI is increasingly integrated into a wide range of software and tools, bracketing AI out of assessments may be challenging.
As generative AI tools such as ChatGPT, Claude, Copilot, Gemini, etc. become more accessible to students, other tools are being developed to try to detect AI-generated text.
However, no AI detection tools have been approved by eSolutions or by the University Data Protection and Privacy Office, and absolutely cannot be used as evidence if reporting a student for suspected breach of the conditions of assessment. Refer to Academic Integrity Detect a Breach for Monash specific guidance.
AI detection tools have not been approved because they raise significant implications for privacy, equity and data security. There is also a lack of evidence as to whether they are actually effective or reliable. AI models are being developed and updated at an increasing pace, meaning that any detection tools are unlikely to keep pace with the latest developments.
Watch: Tim Fawns (Monash Education Academy) and Dave Cormier (University of Windsor) discuss AI detection.

Programmatic, authentic and future-focused assessment View

Programmatic assessment draws upon the notion of assessment for learning. Programmatic assessment involves a series of interconnected assessment methods that build upon each other. Rather than focus on assessing individual kinds of knowledge or skills using a single assessment method, it focuses on assessing broader attributes using multiple methods.

Programmatic assessment supports a holistic assessment frame that is most likely to assure assessment tasks so that they are diverse, authentic, provide relevant evidence of student learning and minimise risk of inappropriate use of generative AI technologies. Research indicates that programmatic assessment has many benefits for students and educators, including increased confidence in assessment decisions, reduced emotional burden and early identification of at-risk students.

Authentic assessment focuses on assessing student learning against learning outcomes aligned with “real-world” contexts and industry expectations (including, but not limited to. work-integrated learning). Authentic assessment tasks require students to apply knowledge and skills in meaningful ways. Designing authentic assessments involves not just focusing on authentic tasks but also on educator judgement of the relevance of the tasks and how they can contribute to students’ sense of belonging to a profession or society. Examples of authentic assessment can include case studies, exhibitions, reflective portfolios, and problem-based inquiries. While some research appears to support the role of authentic assessment (when done well) in reducing academic misconduct, the evidence is unclear and the likely success of this approach will depend on how it is done in its particular context and conditions (Fawns et al., 2024).

Future-focused or “sustainable” assessment advocates assessing the knowledge, skills, and attitudes that students would need throughout their lives beyond graduation. This kind of assessment focuses on developing the ways in which students learn to contribute to their capacities as lifelong learners. A key principle of future-focused assessment is to develop students’ capabilities to make evaluative judgments about what constitutes good work. Some examples of future-focused assessment tasks across a range of disciplines can be found on the Monash University Be Inspired pages.

Each of these assessment elements has the potential to reduce the risk of breaches of academic integrity in relation to generative AI technologies:

by incorporating multiple assessment formats and targeting a range of different kinds of knowledge, programmatic assessment has the potential to capture learning that cannot be replicated by tools like ChatGPT
students are less likely to engage in academic misconduct when they find assessment tasks authentic, personally relevant and more readily applicable to real-world scenarios
future-focused assessment can help students understand how they might learn to work with new technologies like ChatGPT in ways that contribute to their personal and professional development over time.

These programmatic approaches and shifts that are underway as part of the Programmatic Assessment and AI Review (PAAIR) project.

Back: Communication with students about AI Next: Assessment policy and process

Your feedback matters

Let us know how we can improve this resource.

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License

Learning and Teaching: Teach HQ

Learning and Teaching: Teach HQ

AI and assessment

Designing assessment for achievement and demonstration of learning outcomes

Considerations for assessment design

Test current assessment design View

Confirm and coordinate the learning outcomes View

Review the task View

Review the marking procedures View

Contributions

Considerations for assessment conditions

AI and Generative AI tools may be used within educator's guidelines View

Evidence of unassisted human knowledge and skill is required View

Visible evidence of learning and contemporary academic integrity View

Considerations for visible evidence of learning (as an alternative framing to assessment security) and contemporary academic integrity.

Assessment conditions

Programmatic, authentic and future-focused assessment View

Your feedback matters