Tag: #AI

Posted on

Introducing AI Policy to Students

Today we have a guest post from Dr. Devon Cantwell-Chavez, a postdoctoral scholar at the University of Copenhagen and current Chair of the Education and Learning Section of the International Studies Association. Devon shared her take on how to talk with students about AI use and policy on LinkedIN, and generously agreed to repost here on ALPS. You can find her at https://www.linkedin.com/in/devoncantwell/ and @devoncantwell.bsky.social.

Here’s what introducing my AI policy in the first day of class looked like today:

For context, I don’t ban AI in my classrooms because (1) my institution does not allow for this (2) any ban is unenforceable and (3) I’m uninterested in the surveillance necessary to implement a ban. I do tell students I strongly discourage AI and unpack why this is.

Before the class, they were assigned three readings:

– Tressie McMillan Cottom’s op-ed in the NYT on the midness of AI
– the new MIT report on AI and pedagogy
– Kate Crawford’s introduction chapter to Atlas of AI

We start off with basic definition of generative AI that emphasizes the large data corpus, probability outputs, and perceived confidence of those outputs. Then, we look at how governments are applying AI in public sector work around security issues (this is a security studies course). Many of my students will go on to the public sector (or are from the public sector) so I want them to start by thinking about the implications of these tools being in the public sector. These examples include discussing how the UK home office has used these algorithmic mechanisms to impute data for files and to make recommendations for decisions; how the US has used ML based translation software which has led to the rejection of refugee asylum claims due to inconsistencies generated through inaccurate translation; the impact on the public service labor market; and the vagueness surrounding most government guidance on the use of LLMs in work places. Find the slides here.

Then, we talk about AI use in this course. The university requires AI disclosures for all exams and assignments. Those that used AI in this last time I taught the course scored on average about a grade lower (we have weird scoring) than those who did not use AI. We unpacked some of the reasons why this might be.

What did students share and take away from this?
– they are noticing how LLM use atrophies their thinking and can even make them second guess their understanding of an issue or content they are working through
– they notice that the writing suggestions made by AI often make their points and analysis more convulted
– despite observing these things, they worry about “building AI skills” and have heard anecdotes about employers not hiring folks because they disclosed that they do not use AI
– there is a direct relationship between student confidence and LLM use- when they are less confident, they use AI more
– the pressure on students time is a driver for increased AI use
– students said they have not received any instruction on ethics related to AI
– students found the MIT report VERY compelling

We will be revisiting this topic a lot throughout the course due to how AI has become intertwined with contemporary security issues. I’m not sure I will necessarily change student’s AI behaviors but I am hoping through a sustained intervention with this topic throughout the course that students will be more critically aware of AI in future practitioner roles.

Posted on

Meta-Cogitating Motivation, Part 1

As the readers of this blog know, but many of our students don’t, learning requires effort. People who want to learn how to play the guitar, bake croissants, or speak Tibetan are willing to expend considerable effort to learn these skills. Those who don’t aren’t.

Teaching in the age of AI?

At this stage of my career, I expect students to have the minimal level of motivation needed to learn something about the subject of the course I’m teaching. If they prefer outsourcing all cognitive effort to AI because they have no curiosity about, say, how the election of populist authoritarians has led to higher energy prices, they should go somewhere else. My enthusiasm in the classroom won’t by itself magically convert them to my religion.

For my autumn term courses, I will be introducing some exercises that should make the connection between motivation, effort, and learning more apparent to students. One such exercise is a graded* self-evaluation with the following questions:

  1. On a 1-5 scale, with 1 being lowest and 5 being highest, how much of a priority is this course compared to my other courses, work, personal life, etc.?
  2. Given my answer to Question 1, how much effort did I exert last week to learn in this course?
  3. What have I learned in this course?
  4. What specific actions did I take last week to learn more effectively in this course?
  5. How does my answer to Question 4 reflect advice from the “How to Maximize Learning” page in the course’s “Start Here” Canvas module?
  6. How have I connected my experience in this course to one of my interests?

I’ve scheduled this self-evaluation to occur four times across the term. I’m hoping that it will help students realize that learning is like taking one’s mind to the gym. They can either lift the mental weights themselves, or they can waste their time watching other people — or machines — do it. I’m happy to coach them on how to do the former. The latter isn’t my problem.

*Automatic full marks for completing the survey’s questions, courtesy of the LMS/VLE.

Posted on

Iran 2026 Simulation

Immersion in the simulations and games space for the last 15 years has taught me the challenges of creating and running simulations. I’ve written about these before, in ALPS and elsewhere, and time in particular is a barrier to using simulations in the classroom. Time to find or create a simulation, and time to run it in a packed curriculum. Faculty may want to try a simulation, but can’t find one for the topic, skill, or learning goal that they want, or it doesn’t fit the time they have available in their class. The ALPS team has taught workshops on simulation design and we acknowledge the challenges of putting together a bespoke simulation.

Well, AI just solved part of this problem for us. Yesterday, as part of a demo for my ‘Crash Course in AI’, I asked Claude Cowork to create a one-hour simulation on the Iran War for a large, lecture style class that met particular learning goals. In just a few minutes, it gave me a 32 page document that included all the materials I would need to put that simulation into action–everything from learning objectives, set up notes, facilitation timelines, debriefing questions and points to make, gameplay mechanics, and role sheets.

I’m turning this document over to you, ALPS readers. It’s unedited–I have not fact checked it or made any changes, but as a draft, it may be helpful to some of you teaching right now. Looking through it also shows you how powerful a tool AI can be as a teacher partner. Instead of my spending hours putting this material together, I had it write a draft while I was in the middle of a presentation, and now its ripe for editing, adaptation, and changes that reflect my (or in this case, your) judgement. Feel free to use this simulation in your classes–all I ask is that you report back here in the comments or on social media on what changes you made and how it went.

And if you are interested in the prompt I used to get this result, here it is. I had Chat GPT create the prompt for me, too. In the document you will first see a template you can use yourselves, followed by the example of what I used to create etc simulation document.

It’s a new world, my friends. Not sure if its exciting or scary or both, but its bringing radical change to this corner of the teaching and learning field.

Posted on

Six Faculty Archetypes for Navigating AI in Your Teaching

I’ve been developing a lot of workshops, trainings, and short courses on AI lately, all targeted at helping faculty learn the basics of AI and its applications in the college classroom. I wanted to share one part of those efforts that I think will have broad application and help educators who are thinking through their own approach to AI.

While we are all likely subject to institutional policies that affect the role of AI in our classroom, that context does not have to govern our individual perspectives on AI and what role it should play. I’ve talked with faculty who sit on all sides of the AI-spectrum—everything from full-scale denial of the role of AI to those who have thrown themselves into agents and vibe coding. For those that are still figuring out the role they want AI to play in their teaching, though, I have created six archetypes which may help.

Alt text: A 3×2 grid of illustrated panels, each depicting a different faculty archetype. Top row, left to right: a man in a suit labeled “Detective” holds a magnifying glass up to a piece of paper against a blue background with computer screens. A man in a purple shirt labeled “Architect” sits at a drafting table with blueprints and a pencil, with a warm yellow background. A woman in a blue jacket labeled “Pragmatist” sits at a desk with books and a laptop, with students visible behind her against a blue background. Bottom row, left to right: a woman labeled “Coach” sit across from two students at a table in conversation, in a classroom with motivational posters on the wall. A woman labeled “Resister” sits at a desk with a laptop and an open book, one hand raised in a stopping gesture, with a whiteboard behind her showing a diagram of human qualities that AI cannot replace. A man in an orange sweater labeled “Embracer” holds a laptop with a robot icon, surrounded by glowing digital icons against a teal background. Note: image created in Chat GPT and edited by Claude; alt text provided by Claude.

I’m going to walk through each approach and some of the strategies that you can employ if this is the role you see for yourself. Note that of course these are ideal types. You might switch from one role to another based on context, student, or assignment, or see parts of yourself in multiple roles. I share these not to force anyone into a box, but to illuminate the different roles I’m seeing faculty play so that those still figuring this out know what some of their options are.

Continue reading “Six Faculty Archetypes for Navigating AI in Your Teaching”
Posted on

5 Strategies for AI-Resistant Assignments

I recently got to attend the University of Central Florida’s Teaching and Learning with AI Conference in Orlando, and it’s one I highly recommend for anyone interested in practical ideas on how to think about, resist, and use Generative AI in higher education. The presentations are all 25 minutes and the presenters make the most of their short time. I want to share some ideas from one of those presenters, Ashley Evans, professor and Program Chair of Software Development and Cloud Computing at Valencia College. While her field is obviously far from political science, her strategies for transforming assignments from AI-vulnerable to AI-resistant are easily adapted to our field, and worth considering. I want to note that these are my recollections and takes on Evans’ presentation; all ideas here are hers, not mine (excepting the adaptations to polisci). Her website is https://ashley-evans.vercel.app/ and she has told me she’s happy to talk AI with other faculty; contact her by email.

Evans’ stated goals for creating AI resistant assignments include:

  • Reduce student ability to get high marks from solely AI-generated submissions
  • Eliminate the ability of AI to complete the task in place of the student
  • Challenge students to think critically about AI generated output, if used.
  • Relieve instructors from the role of AI-detectives.

All of these seem like worthy goals to me. Most importantly in my mind, she’s interested in taking the weight off of faculty of trying to root out unauthorized AI use, not by banning it, but by revising assignments so that AI is genuinely not useful on the assignment, and by establishing grading criteria that focuses on completion of tasks that AI cannot do well. This is a tough balancing act, but one that I think faculty need to wrestle with as they figure out how much they want to incorporate AI into their classes.

With that, here are her five strategies:

Continue reading “5 Strategies for AI-Resistant Assignments”
Posted on

Quizzes in the Age of AI: an idea from the Teaching Day at ISA

Today is the ‘Teaching Day at ISA’, a free mini-conference held on the first day of the International Studies Association’s Annual Meeting. We’ve got four great panels going, on Creating Compelling Courses; Assignments in the Age of AI; Active Engagement; and Mentoring Early Career Instructors.

So far, the idea that has most stuck with me comes from Dr. Ronnie Olesker of St. Lawrence University from the AI panel. She’s found an innovative way to do reading quizzes that help students engage with the mistakes commonly made by AI. Reading quizzes have long been used to encourage students to engage with the assigned reading, but AI has made it possible for students to quickly generate a quiz-passing summary without actually reading. Dr. Olesker instead feeds her questions into AI, and then counts the mistakes it makes. The quiz asks students to read the AI response and ‘find the mistakes’–with an A going to students who find the same number (or more) of mistakes that she does.

This is a creative way to continue using reading quizzes to encourage student engagement, but also to teach students about the hallucinations and other issues, such as AI’s tendency to leave out important context or information. The quizzes are closed book and in-class, and because they focus on finding errors, students who rely on AI summaries won’t be able to pass the quizzes.

Looking forward to learning more great ideas at this event!

Posted on

Chatbot your Syllabus

I just attended the Teaching and Learning with AI Conference in Orlando, and my brain is brimming with ideas on how to use AI in support of students and faculty. I think most of us remain stuck in the Chat GPT mindset, concerned about academic integrity and students using that platform to write their papers and discussion posts. I’m not downplaying either that, or the privacy and ethical concerns about AI. But that’s just one part of the conversation. In the last two years there has been an explosion of other tools that could have positive impacts in the teaching and learning environment, and as Jeremy noted, we should be talking about those, too.

If you’ve ever had the urge to say, ‘its in the syllabus!’ to a questioning student, I’ve got some good news: there’s an AI for that.

Continue reading “Chatbot your Syllabus”
Posted on

Generative AI changes teaching and learning: how to protect the integrity of assessment

This academic year, the UCL Centre for the Pedagogy of Politics (CPP) is hosting a series of online panel events. Our first event on 30 October was on the theme of ‘Using technology to teach politics’. In this guest post, one of the panellists at that event, Simon Sweeney (University of York), offers further reflections on the challenges involved in higher education’s embracing generative AI, where tools such as ChatGPT call into question issues of authorship and have profound implications for assessment.

A few years ago, we were worrying about students’ using essay mills, a form of contract cheating that plagiarism detection software struggled to identify. The Covid-19 pandemic and online delivery coincided with a reported increase in academic dishonesty (AD). In late-2022 the arrival of generative artificial intelligence (GAI) chatbots like ChatGPT is a further challenge to the integrity of assessment.

Universities realised that banning chatbots was not feasible, as AI has become an established feature in our lives and graduate employment. As educators, we need to respond positively to the opportunities AI presents, recognising its benefits and assimilating AI into teaching and learning practice.

This means developing strategies that accommodate students’ use of GAI while protecting assessment integrity.

Continue reading “Generative AI changes teaching and learning: how to protect the integrity of assessment”