 Okay, absolutely incredible. Even before the 11 o'clock hour here in central time zone in the United States, we're already at our largest liquid margins attendance ever. So clearly this is a topic that folks are interested in and we're excited to have this conversation. Hello, and welcome to Liquid Margins. I'm Jeremy Dean, Vice President Education at Hypothesis. For those new to Hypothesis and Liquid Margins, Hypothesis is a social annotation tool used in classrooms across the world as well as other professional contexts and in everyday life on the internet. Liquid Margins is a show we host in which we gather practitioners of social annotation mostly educators to discuss how they use this technology in their teaching. And you can see we've got some upcoming topics ahead. So follow us to see what we're gonna be talking about in weeks to come. This is going to be a pedagogical discussion about teaching strategies using social annotation. So if you're looking for a basic demo of the Hypothesis tool suggests that you reach out to education at Hypothesis because we're gonna dive in pretty deep pretty quickly here and other ways to get to know the tool if you're new. We'll be taking questions if you have one in the Q&A portion of the Zoom interface here. So feel free to ask a question and we'll have a Q&A period at the end. Looks like there's already a little one in the Q&A section. So we already got some questions here in the first minute of the episode. And if you would like to turn on closed captioning you can also do that through the Zoom interface. So the rise of chat GPT how to work with and around it. Welcome. Today we're here to talk about another technology AI writing tools like chat GPT much the buzz these days. We'll be talking about the role of platforms like chat GPT in education and also how social annotation tools like Hypothesis can be used both with and against AI text generators like chat GPT. For the sake of definition that GPT in chat GPT stands for generative pre-trained transformer. It's an example of generative artificial intelligence which are algorithms that can produce or generate new content. In the case of chat GPT what it produces is text. That is part of what might be called a chat or a conversation. What it's trained on, what it's scraping its words from is the entire body of the text entire body of text on the open internet. Here's how chat GPT defines itself when prompted. And what's remarkable to me and I think to many others speaking as an English teacher for many years is that chat GPT produces reasonably well-composed writing when prompted. It's often wrong and it's prose can be stilted but it's pretty good for some it's scary good. So the advent of chat GPT has prompted quite a lot of human writing and conversation and chatter over the past couple of months about what this technology means in education generally but also what it means in the English classroom specifically where a major part of what we do is teach students to write or if you will train them as humans to generate text. There's a lot of fear about students using chat GPT in place of their own writing and thinking about the use of AI for grading about the need for teachers and students to teach and learn to write at all if computers can be so fluent. It is indeed existential if you've been reading the articles published on this. There's also a lot of curiosity and self-reflection. If computers can write like this what is the job of the writing teacher? How have we been teaching? What have we been teaching? How should we be teaching in this new age? A long time fellow traveler of mine in writing and social annotation circles Dr. Mia Zamora at Keene University in New Jersey who somehow was already teaching a class on this topic expresses what I think a lot of us are thinking and feeling nicely in her course description for that class. What does it mean to write in the age of AI and GPT-3? The algorithms can write faster than we can. AI is now a mastered structure form and syntax. Still it doesn't have a clue about meaning. At this stage you can offer a prompt and the program will spit out something useful within seconds. For those will be grammatically correct. Words will be coherent. It knows basic paragraph structure and it connects ideas together effectively. So where does that leave us human writers? Today we're joined by four English educators and users of hypothesis social annotation to try to answer some of these questions. I'm gonna let them briefly introduce themselves before we dive into it. So let's go left to right here starting with Kat who isn't called out as an English educator in this current slide but I know for a fact has taught English in the classroom. And so we'll just go one by one here and we'd love to hear where you teach, what you teach and what kind of courses and Kat why don't you kick it off for us please? For sure, hi everyone. Thank you so much for being here. Thanks Jeremy for having me. So I'm Kat King. I teach freshman comp and critical thinking Las Pasitas College which is a community college in California and I'm a project coordinator at Diablo Valley College where I do a lot of training on instructional technology, supporting faculty in online education and incorporating technology into their classrooms, talk pedagogy, all that kind of stuff. So excited to be here talking with you all today. Go to you Joel next. Hey Jeremy. My name is Joel Glenn and I am an assistant professor at the College of Western Idaho we're a community college, semi world community college in the Treasure Valley of Idaho. And I teach first year writing, American literature, a few other courses but those are the main ones. And I also facilitate a little bit at the Rebus organization and their textbook, their OBR textbook program there. Yeah, so I'm excited about the open world and HPT is kind of an interesting kind of intersection with that. Thanks for being here Joel, Rachel go ahead. Sorry. Yes, hi, I'm Rachel Rigolino. I've been teaching at the State University of New York at New Pals and I did the math for 30 years. This is my 30th year. I work primarily with developmental writers but I also teach business writing. So this is really interesting to me and I'm a big ed tech nerd, serve on all kinds of committees about educational technology. So yep, and I'm excited to talk about this too. Excellent. Finally, Nick. Hi, thanks for having me here, Jeremy. I'm Nick Lollardo. I teach in the expository writing program at OU which we teach second semester freshman composition, small seminars and my background is maybe relevant insofar as I'm an American lit professor and originally and my particular interest is American poetry, especially experimental poetry. And so I look at, but generated writing from that perspective as well. Excellent, glad to have the creative writing perspective represented here. Well, great. Let's get into it. First, I just want to hear from you guys general and maybe we'll go in reverse order here. What's the mood like on campus these days with chat GPT, both with your campus colleagues but also in your networks. We'll start with you, Nick. I know you've done some presentations at the University of Oklahoma on this topic already. So how are folks thinking and feeling about it at University of Oklahoma? Actually just one, but I mean, I appreciate the shout out. I presented at an academic tech expo right before this semester started. So I guess I pitched that presentation in December when the buzz was already happening, gave the presentation mid-January. A bunch of people came out for the topic and they had questions from all different perspectives, all different kind of disciplinary takes, discipline related concerns. So there was a lot of buzz, there was a lot of concern and I think people also still have very different reactions as well. Rachel, I know you've written about this for inside out higher education, is that right? Yeah, and I was being a bit provocative in that article but to just piggyback on what Nick was saying, I think it depends on my campus who you're talking to. We had a great presentation two days ago in the faculty center, a person out of visual rhetoric that was really super interested in it, thought it was great, not great, just categorically but really energized about it. But then I came out of an English department meeting and it's like we're going back to oral exams and handwritten in class tests. So there's a real difference and diversity of opinion, certainly. That's fascinating. Joel, how about in College of Western Idaho? Yeah, there's also a similar like divergence here. I would say most faculty are fairly excited about it and I'm on the arts and humanities side of things and we actually, I'm so a part of the English and actually surprised by what Rachel reported at her institution, but like we're mostly excited about what it has to offer and like, I mean, I've seen it, some of the conversations have kind of viewed it as potentially an equity tool depending on how it's used. And I tried to try to play around with that a little bit. But it's also like there's, you notice kind of speaking of equity, faculty who traffic in certain networks have a different, like it does a different vibe with them, surrounding chat CPC versus others who are like, hey, we're just trying to do our job. And things have been working so far, let's not move too fast. And I think the pacing of how this rolls out, I noticed this creating a little bit of uncertainty and friction in some places. And that's kind of what I'm noticing. Yeah, so I would say that's probably one of my concerns is just the pacing of things and how to adjust. But yeah, otherwise overall like fairly positive. They are as promised, moving fast and breaking things. Kat, what's the vibe like at Diablo Valley and Contra Costa? Similar to what I'm hearing here, this resonates. We've got everything from faculty who are very curious and excited to engage with students about this. And then it sounds like as other places, we've got instructors like having this existential crisis. Like this means we can't do online education anymore because students will cheat their way through the whole thing. And kind of wanting to have that surveillance state moment like we saw in the beginning of the pandemic with the rise of like Praktorio and Honor Lock and watching students as they do everything. So there's a little bit of the whole spectrum. I think there is a lot of excitement though and I'm happy to see that because I understand that sort of existential crisis as this is profoundly disruptive to teaching. But in my mind, like it's students who are the ones who get to have the existential crisis here not faculty for the most part. We're in an industry where we have like tenure and strong unions and shared governance and like things are gonna be slow to change in some ways. But for students, I think once we get past the, oh wow, this is the shiny new tool that is maybe a hack to education. I think there are gonna be some profound moments of potentially despair as they realize how this type of technology potentially is gonna disrupt industry and like what they thought they wanted to do with their lives, the things they're studying. Suddenly you have a student who is like, your creative student going into artistic stuff is looking at bots that can generate some pretty impressive art. You have practical students who are going into computer science that are now realizing AI can generate code really quickly. I think there is gonna be a lot of anxiety as students start to realize what this means on a larger level. And I'm hopeful and see a lot of faculty stepping in to support students kind of navigating that conversation and what that means. That's great. Well, there's quite a range of topics to dive into here. I'm super interested to hear more on the equity tip, Joel and others. And also obviously we mentioned plagiarism here, which is definitely one of the words that's getting used a lot in these conversations. There's a lot to talk about. Let's use Kat's last topic there in terms of students. Have you guys talked to your students about this? Are they talking about it? What's the student perspective, if any at this point? I'll jump in if we're moving to a more informal mode here. Just to say that I introduced the bot and the implications yesterday in the context of introducing my initial essay assignment for a course. And as it happens, the assignment was a manifesto to write from a generation Z perspective. The classes theme is generational conflict and media generations essentially. So I showed them an essay, which I said had good and bad qualities. This was a true statement on my part. The essay was written by chat GPT. I temporarily withheld that information, got some student feedback first, then revealed the essay. The students were quite critical of the essay in what I thought were generally very perceptive ways. They commented on a kind of a generic template, lack of detail, a kind of one level of abstraction over the whole surface of the text, no particular feeling of authorial identity, and a refusal to take a strong stand, which was particularly ironic in the context of a manifesto assignment. Somebody said, this is a bland manifesto. Blandness is not a virtue of that genre as traditionally conceived. So when I then showed the students the bot version of the essay, rather than making the plagiarism approach, which is kind of both obvious, and in some ways insulting to them, I wanted to make the point that like, don't worry. The bot's not gonna replace you, you are human beings with a need to express yourselves, and that's gonna remain the case, at least for the foreseeable future. So that was how I was able to introduce it this semester. Go ahead, Joel. I think you're muted. Thanks. Something Nick brought up, but I thought that's usually smart of him is that Nick showed generated writing, had them critique it, and then revealed like, hey, this is generated by a chat to BT, so let's talk about does that make a difference? That was kind of, that's a smart way to go about it. Something I've noticed is, and I just ran across this morning when I was working with a student using chat to BT to leave feedback on an essay, where you're kind of playing around with how it might be useful, but also how we need to be critical thinkers when we respond to this feedback. And something you'll notice if you play around with like feedback or essays that involve citations, it's actually quite, can be really clunky with citation issues. And the student, like I saw you ask the student, like, hey, what's your impression of this feedback so far? And she said, hey, I thought it was really good because I pointed out my citation issue, you have rewrote the paragraph correcting for the citation. And I looked at it, it was actually wrong. The citation wasn't correct. And I was like, hey, this isn't a good example of where, of why, when you're in course like this and you wanna use a tool, you need to understand the limitations of chat to BT and other AI writing tools and how they can, you can be productive with them, but also how they will hallucinate, right? Just kind of tell you things to make you happy that aren't true and just kind of get things just flat out wrong, right? So that's, I think something that I'm trying to work on right now is figure out how to coach students and how to use the tool. And as we kind of work through using the tool in class in real time, talking about the feedback they're getting and like my student, I noticed they just trusted the chat, the chat to BT, because it was so convincing. Like, oh, like surely this must be right. And there was no, like, oh, it can be, like there was no thought that it could be wrong. It just felt really convincing, yeah. Yeah, so there's not more to say, but I'll stop there for now. Reminds me of the early tidal debates about using Wikipedia as a source. Joel, I just wanted one point of clarification before we hear from Rachel. Did you say that you had a student enter an essay they wrote in chat to BT to get feedback on the essay? What, how were you using feedback there? So this week and one of my courses, we've started using the tool and that in itself, like we can talk about the issues with trying to like, like walk students through just logging in is an issue. So like one of the questions is about concerns related to equity students who struggle or unprepared for certain kinds of digital literacy. It'll take them longer just to get set up properly on top of that chat to BT. Is the server is frequently busy when you try to work through this synchronously and that's a problem, like how do you plan for that? And so I've had to actually tell the students, like, hey, we're gonna work for, we're gonna try chat BT, chat GPT. We're gonna use it to practice getting feedback on small parts of our writing not entire essays just yet. And we're going and, but here's a backup link. If that doesn't work, we're gonna have to go to the opening a playground and play around with that, even though it's not as good or as intuitive as a chat GPT. And so yeah, that's kind of the process we work through and it's a frequent issue. And this is part of like, things are moving very quickly. There's one question is just about the pedagogy. Let's assume everything works as we expected to. Like there's a lot of pedagogical issues, but there's a whole range of technical issues of how to make this work in the classroom right now because it's in flux, like everything's in flux, right? Yeah, for sure. Rachel, what's been your feedback from students or have you introduced this to students? No, and I'll tell you why I briefly mentioned it and I've gotten some sense of who may be using it, not among my students per se, but one thing I wanted to say it, Joel said, things are going so quickly. The reason, and this has to do with equity too, the reason I'm not playing with it in the classroom as of yet is this whole kind of reaction to it, which is so negative to the point of people putting in their syllabi, if you are caught using this, you will just fail the course. I mean, I'm serious. So I don't wanna introduce it and then have my students, oh, you've taught them to use this, they're cheating. So I'm kind of, usually I'm pretty impulsive about ed tech, I just wanna use technology right away, but I'm kind of stepping back a little bit because I kind of wanna protect my students and myself, till we see what's going on. So next question is just about how you've adjusted if you've adjusted or done anything differently. It sounds like you're waiting and thinking, Rachel, but maybe you have adjusted in some way. How have you adjusted your course, your preparation, your assignments, your activities for this moment? Have you, I mean, I know we can maybe start with you, Joel, because I know you have course policies, right? That you've written up around chat GPT. So the question is around how you've adjusted your course setup in various ways, if at all to address the fact that chat GPT is here for spring 23 courses. Sure, yeah, that's a great question. So I think, so just a little bit of background here, I was playing around with or introducing open AI, the playground version that was in beta, which still seems to be in beta in the summer. So it's been around for several months, that version of it in earlier iteration. And one of my courses was like, hey, you know what? I think this is gonna be big. Let's start playing around with this. We used it very informally, kind of playfully, we'd kind of workshop some things. And then later on in the fall, chat GPT rolled out and it was clear that this was being picked up by students at a number of institutions. We started to get reports by faculty, okay, I'm noticed this was AI generated, I failed the student, and then there's like this lot of discussion about what the right recourse is when we think a student is cheating. And so coming into the spring semester, I just saw suggestions from a number of people about Sylvester language. And so I decided to design around, like design with chat GPT in mind for this semester, but like in a very basic way. And I wanted to provide some kind of structure and framework for students to use it productively and not like on the sly. And so this, I think kind of like, conversation with Rachel is concerned here, I felt like it was incumbent upon me if I'm gonna introduce this in my course, I need to provide a framework. And also I need to take on the responsibility of I'm introducing technology that, it seems like 80% of these students are 90% did not know about until I mentioned it in the syllabus. And so what do I need to include there that starts them down the right path and reduces the likelihood they'll use it to cheat. And so the syllabus language includes things like kind of a brief guide, like here are problems with chat GPT, here's what hallucination means. It's not hooked up to the internet in the way that you think it is. Just all the downsides, I tried to emphasize, but then try to clarify how we would use it in the course and then how it related to academic integrity. And I know that's a kind of non-growing conversation, but at least that framework provided me with a structure I could go back to. So it's not like, hey, by the way, week four, you all heard about chat, let's start playing with it, right? Like they know there's already some guidelines in place. Does that answer your start to answer your question, Jeremy? Yeah, and I'm curious for others, you guys have been all very good about raising your hands. How have you or how have you seen others adjust their courses and their activities to deal with chat GPT? I'm happy to jump in here. I am trying to lean in and start the conversation with students, and I think just how quickly this happened really reinforced for me like that I need to weave in digital literacy skills into my course to again get students thinking about, you know, information they're gonna encounter on the open web, and what does that mean when that information is potentially generated by AI and all of those things. I have been trying to lean in a little bit and have, because it kind of disrupted what I thought I was gonna do this semester, I started playing around with chat GPT to see how it could help me. So like for example, I went into chat GPT and said, write me a lesson plan about, that gets, you know, first year college students thinking about chat GPT and what it means for them. And it actually produced like a pretty decent lesson plan that started with like, okay, you start with a little direct instruction similar to what you did here, Jeremy, kind of defining some terms. And then here's this article that students can read and annotate together, and then they'll break out and have a discussion about it and discuss it in small groups. And then as the instructor, you're gonna pull everybody back together and talk about it. And I was like, oh, okay, this is interesting. And so I started playing around with like, what else can I do? And I'm teaching a novel I haven't taught before. It's our campus read at Las Pesitas. It's there there by Tommy Orn. So I was like, hey, give me a handful of essay prompts. About this novel. And I was skeptical at what it could do. Five seconds later, it gave me like seven essay prompts that were all like pretty decent. I was like, wow. So then I was like, okay, now, now write me a model essay that my students can see on these essay prompts. And bloop, it wrote me out model essays that I could share with students, which usually I'd have to wait a semester to get those model essays to share with students. So I've been trying to kind of lean in and see where can I, this is kind of disrupting my time and shifting my instruction, but how can I help have ChatGPT help me to the left, if that makes sense. Other thoughts about how they, how you all have adjusted your courses or your teaching style for ChatGPT? I think a lot of folks are still at the beginning of the process, but if I could just say something, I mean, potentially this is the kind of adjustment that will be most beneficial if it happened at some kind of campus-wide or at least programmatic level. And I say this as somebody who has been involved in WACC style work at university writing across the curriculum, writing enriched curriculum. And maybe a little pessimism is warranted at this moment, which is just to say that for a long time, writing instructors, writing professionals have sought to kind of spread the gospel of better writing pedagogy across campus. And for a long time, they've struggled to get a foothold for a long set of reasons that, we have no time to go into here. But I do think, this potentially presents a kind of opportunity if people in writing pedagogy are seen as being able to provide valuable answers or to help their response to a sudden crisis. So I'm certainly thinking about what I might, what more I might want to say about this spot and the problem of bot-generated text in general, at a university-wide level. But I would also say that to some extent, the answer is an old-fashioned answer which has already been offered and rejected, which is to say writing assignments need to be made meaningful to students in certain kind of ways and info dump exam question style writing assignments are not going to be helpful. That if you're teaching writing solely as a means of information transfer, you need to sort of rethink why you're using writing at all for that particular process. Did that come across as feisty? I'm not sure yet. I love that. I think that's great. And as you've got some hands up from your colleagues here, so we'll start with Rachel and then go to Joel. And I would add, yeah, I agree with those type of exam quite, I mean, come on. But to get to this existential crisis, so you're talking about writing is critical thinking and that was our discussion in the English department. I mean, people hold this very dear that when you're writing, you're engaging critically with a text, right? So the idea that students, the English majors will be using chat GPT. And see, I teach mostly non-majors, you know? So I was kind of appalled to think, I mean, this surprised me to hear my colleagues go on about our majors. And I think that exactly if you're invested, a student who really is invested in an assignment, it's not going to just run to chat GPT and type in the prompt and, you know, yeah, I was a little bit surprised by that, but the connection between critical thinking and the process of writing, I mean, all of us English teachers, yeah. So. Joel, you had your hand up earlier. Yeah, it's so like the Rachel's comment about like the importance of critical thinking is resurfacing, right? That's always kind of lurking in the background. And I think what's interesting about the chat is that I would imagine what critical thinking means in each course. We're sorry, what critical thinking means is going to be determined by the context of the course in which the chat GPT is deployed. Like, it'll kind of like raise certain questions about like, okay, what critical thinking skills do we teach in this discipline and this course that would allow us to evaluate the usefulness of this generated text. So that's going to be highly relevant or that's becoming a highly, highly relevant question. But thinking about the assessment piece because that's part of the anxiety, I think I've seen less anxiety among faculty and this does privilege. Maybe this privilege is certain kinds of disciplines over others, but maybe not. But I've seen faculty who have been shifting towards process-based assessments and reflective writing becoming more comfortable with this. Whereas those who like had not yet implemented those practices or yeah, you know, or it just doesn't play as much of a role in the course. And that makes sense. Like it's going to be all over the place depending on the field that you teach with them. That's where I saw war anxiety. So this is anecdotal, but that's kind of like my observation there. For me, the importance of process-based assessments and then students reflecting on and analyzing in the form of cover letters or other kinds of writing where they're actively rethinking about how they're using certain, the choices they're making, how they're using certain tools to achieve a certain result, that's very comfortable. That can become very comfortable with deploying tools like this. So like I'm clearly like on the less, like I'm kind of, I'm playing more, I'm maybe I'm moving a little bit faster than others or I'm more comfortable with moving faster. There are risks to that. I'm breaking things that I need to be careful about. But I think where the kinds of assessments that we had already been moving towards in my English department and that I've seen elsewhere like allows us to kind of pick this up a little bit faster. That's a great opportunity, I think, to transition to social annotation and other strategies and tool tech practices that we might be able to use to alter the way we teach them. And you guys are all veteran social annotation users. So I would just be curious about the role you think social annotation plays. Joel, you mentioned I think a key piece of this which is the process oriented pedagogy versus product oriented, right? If there's a simple paper assignment that's been used semester after semester, it's pretty easy to, and that's kind of the culminating summative thing. It's relatively easy to plagiarize that. I mean, if that is a concern and it's also just not maybe the most interesting type of learning then maybe students would be less invested in it. But I just be interested in y'all's take on, if you see social annotation as a way to, I guess at first, rethink what we're assessing and what we're assigning. I don't know who wants to take that one up first. In ways that might be sort of an anecdotal to potential abuses of chat, GBT. Go ahead, Kat, sorry. Yeah, yeah, I didn't know. So this is interesting. Just again, a lot of this resonates. I'm a longtime user of social annotation and have been a big fan for a long time. I recently changed the way I waited my grades in my class where I used to do a lot of social annotation but these were kind of more formative, quick check-in assessments and then the bulk of students grade was still like that essay that the annotation helped prepare them for that and so it was kind of a smaller percentage of their grade and then the essay was this big thing. I'm now shifting the way I wait my grades and really thinking about like the work that happens in the margins, those annotations, those engagements with the text as being probably in many ways a more accurate reflection of how students are thinking critically and understanding content in class. I think that annotation is really important because as you get into these conversations about all our students using chat GBT to cheat, I think it's good to pause and ask students like or ask yourselves like why do students cheat and ask students like why did you cheat, right? When you come across that and for the most part what I've heard when I have had to have those uncomfortable conversations with students is like, oh my gosh, I was so busy, I've been working and I fell behind in the reading or I wasn't understanding the reading. There's like so many students who have undiagnosed learning disabilities like dyslexia or a student with ADHD who was having a hard time focusing in on the reading using social annotation to like help keep students on track because you can see are they engaging with the reading, are they keeping up with it, are they trying, is there an area where I can see, oh, the students trying but there's a misunderstanding here I can address in the moment and it's not like that student gets six weeks into the course and it's so long that then they're feeling like they need to go and use chat GBT. So really like leveraging the power of hypothesis to provide that like consistent ongoing feedback loop with students to kind of check in and help make sure they're supportive in what we're asking them to do. I love it. Rachel's got a hand up, go ahead, Rachel. Yeah, just said again, Kat, that was great. I mean, the critical reading piece is always the stuff that that's how I came into using hypothesis. Two years ago, we did an internal assessment and realized the students critical reading skills. I mean, we knew they weren't strong but you could see with the help of hypothesis which we piloted, it did improve. Now we had some guided annotations, guiding them through the reading, but with chat GBT and that kind of was my thesis in my article on inside higher ed, much of what we're gonna train students to be are very critical readers, like yesterday with Bard, a huge error, right? I mean, it's embarrassing, right? They launch it and then there's this error factual error. So information literacy and using social annotation to look very closely at scholarly journal articles at any kind of writing. I think it's just brilliant, it works so well. I'd be remiss if I didn't mention I sort of take issue with the use of Bard for their new search, that's our word in English literature, I feel like. Other thoughts about the role social annotation can play in changing the way we teach to address this chat GBT moment? Go ahead, Nick and then Joel. Okay, just quickly then I will say, I think for me the global reason, like the overall reason to use annotation and I don't think any of my particular practices are super distinctive, but from the beginning of a class, I try to suggest that annotation is gonna be the concrete bridge between reading and writing and that I'm trying to get them to read as writers, to read as potential writers, but to start actualizing that potential by writing in the margins of texts. And so I think waiting that kind of writing more heavily in how I evaluate students seems like an obvious direction to go in, but I also think that that kind of writing has the effect of, there's no reason. If one annotates on texts, that's also just a concrete way of getting students to do the reason, to do the reading if I wanna just say the pragmatic thing that I know some people will need to hear. Requiring annotations with a rubric is obviously a really effective way of getting students to pay close attention to texts. And in evaluations, I have consistently had students say that engaging with the details of paragraphs was not something that they were accustomed to having to do in college and that they were kind of shocked in an ultimately salutary way at having to do that. Joe, go ahead. Yeah, totally agree, Nick. So I dropped in a couple of links into the chat there. I think one, so on the student-facing side, one thing that I'm trying to think about right now is the role of hypothesis and helping students, or helping coach students and prompt and better prompts and how to prompt, right? That seems to be like a key practice surrounding these AI generators is understanding what is involved with the prompts and how to make them as effective as possible. And I think a hypothesis and other social annotation tools could be really, really effective here. I'm imagining situations where the instructor might develop a series of conversations with chat GPT and then how students, let's say it's an asynchronous course, how students who don't have the luxury of doing that real-time kind of coaching, how students respond with social annotation on the conversation, noticing what's working, what's not, exercises, critical thinking skills within the context of the discipline about why certain generated texts might be useful and other texts might not be. So two things that are related here. One is, I sent two links in the chat. So you might be able to pull up that first link, but or you can just share it either way, Jeremy, but that's a link to the open AI guide to educators or it's called educator considerations for chat GPT. And this, if you log in, I sent the link that has the Viya.hypothesis routing. So it kind of populates that hypothesis pain. Some of the people attending may not be familiar with that. So I don't know if it's gonna automatically populate or not, but if you do sign up for hypothesis and you log in, you'll see comments in the margins. And this is an example of how educators are using hypothesis to communicate with other educators about concerns related to chat GPT on the open AI website. And Anna Mills here is the one that is commenting quite a bit. She's really active in this area and it's really on top of things. And I asked Anna a question in response to one of the things. Yeah, there you go Jeremy, thanks. So this is, you can see on the right hand side how Anna Mills is kind of annotating open AI's document educator considerations for chat GPT. And this is like their vision of how educators might should use their tool, right? And so Anna, I think she was invited actually to leave feedback on this document before it was actually published to the public. And so she's going back in and annotated and expressing her concerns. And also some areas where she's glad to see the progress they're making. Someone asked in the Q&A about citations, how to site chat GPT, this actually addresses that. In part, you'll see a code, you'll see a code for bib sex that's kind of mentioned. And I'll just say one more thing here. So we have this really cool use of socially annotations for educators to be connecting and discussing what's going on surrounding this technology. If you want to socially annotate chat GPT conversations, one tool that might work. And I haven't done this yet. So I can't promise that I will, but that's share GPT. And that is a Chrome extension. And what happens is that when you have a GPT conversation, you click on the Chrome extension and I'll generate an HTML link to that conversation, which you could then feed into hypothesis. So just a little thing there, you might use that as like a citation method as well until OpenAI actually rolls out their stable HTML links to these conversations. I think the guide says they're in the process of doing that, but it's not quite there. Just want to call out a couple of things here. The way I've been thinking about this is, maybe this isn't how I've been thinking about it, but let me rephrase this. How might a traditional English educator not particularly interested in chat GPT using it as part of their course, use certain teaching strategies or assignments or things like social annotation to rethink how they teach traditional English composition or English literature. And I think we heard from Nick and Rachel and Cat about the importance of critical reading and how social annotation emphasizes that. Cat also made a really good point about when social annotation is part of the process, the teacher is more present and the student is getting more support. So maybe that leads to less of a situation where somebody would get up worried and enter a prompt into chat GPT and then submit the output as their own. And Joel kind of pushed us forward to thinking about critical engagement with and social annotation with chat GPT. So just to sort of, for everybody's sake, there is the use of a hypothesis to annotate the site for open AI. And it's great and I'm glad you called that out. I'll drop a link in or I dropped a link in some folks are using chat GPT EDU as a tag with hypothesis to annotate not just open AI, but I annotated your article, Rachel, from Inside Higher Ed with that hypothesis and with that tag and I've annotated Anna Mills's work as well. So I definitely recommend engaging with the debate yourself as a scholar and a teacher. And perhaps if you're teaching about social annotation, having students annotate with hypothesis on the web, there's a great set of articles that Anna's put together that I'll share in a second. And then the other piece that Joel mentioned is share GPT and I have used this, which creates a stable URL for a session with the chat bot and thus allows in terms of being critical consumers of technology and helping our students become critical consumers of technology, they could have a session on a topic with chat GPT and then annotate it in terms of where its mistakes are, where its biases are, where it's insights are, where it fails to properly cite its source, where it hallucinates as Joel said. So share GPT is a great way to use hypothesis to annotate a session with chat GPT. So we've got about 10 minutes left, but I'm interested just in hearing more from y'all about social annotation kind of to combat traditional teaching strategies that might be problematic for chat GPT, but also innovative ways to use social annotation with chat GPT. We've already got a taste of that, but Rachel, I'll see you have your hand up, so go for it. Just really quickly, I think I need to take a page out of Joel's book, he seems more relaxed about all of it. It was kind of brainstorming things I could do. Right, I plugged in a scholarly journal, like part of a random scholarly journal article and I said, put this in a silent form. And it did it. I mean, you know, so I'm like, we could have so much fun with this, you know, like let's put something in and make it sound like Ernest Hemingway, you know, in my short story class and then we can like take that text and we can decide, you know, annotate it. Where is this Hemingway-esque? So I think there's, you know, with some imagination and I just, as I say, I'm just fearful. I guess as I get older, I'm a little bit like, oh, you know, 20 years ago, I would have been gunk-ho and pulsive, let me go with this, but I think there's a lot of great stuff that we could be doing with pairing these up. Anyone else? Anyone else gonna follow Joel into the world of diving in with the chat GPT and the other AI tools and being critical and thoughtful as we do? Go ahead, Nick. No, I think I am, you know, in the near future, picking up on Rachel's point, I think the extent to which chat GPT can potentially be used, you know, to draw students' attention to the texture of text, right? That it's not just an information bearing medium. There's more going on here. Then annotation-based assignments are a really good way of getting anyone, you know, to focus their attention on that aspect of textuality. And asking students to generate an essay themselves and, you know, then to mark it up using hypothesis, you know, with a particular rubric, for example, to serve as the critic, that seems like an assignment that would both integrate the technology into an already existing pedagogy, but not necessarily take the full dive into, you know, against originality, and we must all write conceptually in the future. Though I myself might be temperamentally willing to take that dive at some point, I think a lot of these questions, and maybe what I'm saying now connects up to the earlier conversation as well, I do think in a lot of our answers, we need to remember that the solution, sometimes when we're talking about solutions, we end up restating. And so the instructor of record needs to spend more time and do more work here. And, you know, the instructor of record, if they're teaching writing, their labor conditions are often not optimal. And so the tension here, one of the things about web annotation that I like so much is it's a persuasive way to use technology to still teach in what I would call kind of a traditional way, potentially. Rachel, I may be misremembering, but was it your article on Inside Hero that talks about editing? Yeah, so what happened to me very briefly, I was fired from my side gig, which was writing for content-driven advertising for a car company, where I wrote about things like where do you, how do you replace a battery in a 2012 Cooper Mini or something? So I was doing this writing and I was giving 20 minutes to do it for each article. And of course I was replaced by a chatbot, right? So, I mean, of course, I mean, it made sense. So that's where I got into this thought of being thinking about that about, well, my business students, they're gonna have to know how to do this. I mean, this is kind of some of the writing they're gonna be doing will be then to edit. And then my editor is then having to look at, is this stuff correct? So I think the shift, that was kind of my argument that the shift for many of our students, not our English lit majors maybe, but for many of our students will be this role of editing what chat GPT type of AI produces in the workforce. So yeah. I thought that was a really, really interesting point. Well, we have just a few minutes left. I just want to give everybody a final word to maybe tell us where do we go from here? What are your thoughts? Again, I think there's a lot of anxiety out there, although you guys all seem pretty relaxed. But I've sensed a lot of anxiety. There's a question about, we haven't really gotten to the Q&A effectively, I think, but there's a question about where the anxiety comes from. And I guess my question for you to kind of riff off of that is where do we take the anxiety from here or on chat GPT? I can jump in and just say, I think like as much as we can channel that anxiety into curiosity, I think the better will be. Like I'm sitting here writing down notes. I think there's brilliant ideas that I want to go now try. I'm excited, yeah. Yeah, and I can see why this feels like this crisis moment. Because what does it mean for our role as teachers or what does the future of education look like? But there've been big disruptors to education before the calculator, COVID, like a million things and to be honest, this isn't so crazy. Like if our students really wanted to learn any of the content we teach them, there's a million YouTube videos and the library, they could go and like learn all that stuff on their own if they were super motivated. I think it kind of reinforces our role as educators as helping guide students through content, cheerlead them on, encourage them and through community building tools like social annotation and hypothesis, we can still build that community with our students and help them kind of understand this new world and I'm excited for going forward even as strange as it is. I think there's a lot to be excited about. Thanks, Kat. I'll just finish up here going, Joel, Rachel and Nicola have the last word. Thanks, Jeremy. Yeah, I'm looking at the chat here and notice like some people are asking or expressing concern about like, hey, you know, I'm at a more foundational level. I just wanted to know how to get started with chat, CPT, this, you know, some of this conversation might be higher level or at a later downstream of kind of where I am. And if others, you know, like who are attending and have suggestions for what personally allowed help them get started, like, you know, feel free to drop that into the chat. I will say like, you know, I feel free to reach out to me or anyone else on this panel for, you know, more help. I'm open to workshopping with anyone if you want to talk about this. And I think I just want to put this, I want this to be my final comment, which is what I appreciate about this disruption because it's kind of been that to a certain extent. What I appreciate is how many conversations it's provoked and how many people I've talked to recently about what they're doing in their classroom and webinars like this, hypothesis conversations, Twitter conversations, Facebook conversations, like it's generating a lot of important conversations about our vision for the next, you know, five, 10, 15 years. And, you know, at the very least, that's a really, really good thing, I think. And just quickly, I would say about what Kat said about who we are as instructors, right? And social annotation to me helps more than even a discussion board. And I used to love discussion board, but they're kind of stilted. But that relationship that you build with students, I think, you know, it's part of this anxiety about the students cheating and this and that. But building relationship with students, I think it's just so important and it even speaks to some of this anxiety about the unethical use of technology. But it's always going to be at the core of teaching. I mean, at least the kind of teaching it seems that all of us are doing. So I'm just throwing that out there. Okay. Yeah, well, let me pick up on a couple of these. Wait a second. Did I interrupt? Are we last? Am I last? A couple of points that my fellow panelists here have made disruption is absolutely going to happen. And social annotation, you know, preserves traditional reading practices or remediates traditional reading practices, you know, in the context of digital technology, which I think makes it such a versatile resource. But I was doing my own reading for last week's class and remembered a story that is told, was told about the First World War, famously, told by Groot Groot Stein in this case, where she writes that the generals had all planned to fight a 19th century war using 20th century technology. And then we know what happened. A 20th century war is what happened. And to some extent, we're in the academy, technology is going to be ahead of us. We can respond, but we cannot anticipate so that our responsiveness ultimately is just gonna be crucial. Well, that's wonderful. I really appreciate this conversation, y'all. And thank you very much for joining us today. As Joel pointed out, the conversation continue online on Twitter. People are also using Hypothesis to annotate. And if you're here and your school is not already using Hypothesis, we'd love to have your school join us. We have a summer booster promotion right now. It's a heavily discounted pricing for unlimited summer usage, includes implementation, training and certification through our new Hypothesis Academy program. I don't know if you guys are certified Hypothesis educators. Certainly you're advanced Hypothesis educators in my mind, but now we have a certification. The summer is a great time to get started with Hypothesis and increase engagement and try it ahead of an adoption in the fall. So, and there's also if you sign up for multi-year deal we'll apply the cost of a summer booster promotion to a multi-year deal prescription at your school. So reach out to education at Hypothesis if you're interested, please. And again, I just wanna really thank y'all for all your work before we met today and had a conversation. Now you all are interacting with colleagues and with the wider network of conversations around this issue. And I think it's really important to have leadership, thoughtful leadership like that and commend you for that. And also thank you for taking time to join us today. To the audience, thanks for coming. Please join us on a future liquid margins. And in the meantime, go forth and annotate. Thanks everybody.