 We are recording, Chris. Good morning, everybody. Good morning. Hello. Hopefully you can hear us. I get to hear how horrible and tinny that theme tune sounds as well when joining. Well, one day, one day, we will have that piece of kit that we've been lusting after, something like the Roadcaster Pro, where we'll be able to mix all the different channels of audio in and get all the little jingles. You say we. You say we. We say we. Well, I mean, ultimately, if you can, well, if you could give me the money and then I could own it, I think that would be the ideal. Anyway, this is enough of this. My name is Chris Morrison. My name is Jane Secker. We're the cool sync chairs. And this is the 64th, or is it the 63rd B webinar that we're running? We had a discussion. I ran that past. Yeah, I ran that past Alex. Alex fully endorses the decision to make it 64 as a separate entity in its own right. OK, excellent. So yes, the cool thing is the Copyright and Online Learning Special Interest Group part of the Association for Learning Technology. We've been running these for some time now. And let's talk about what we're actually going to be doing in today's webinar. Yeah, we've got a couple of bits of copyright news, haven't we? We have a couple of bits of copyright news, keeping it quite brief because we want to allow plenty of time for discussion on the very important topic of copyright and artificial intelligence. And we will have Alex Fennellen joining us again. I believe he's been here for a while. It was me that had the problems connecting this morning, wasn't it? Yeah, absolutely, yeah. Hello, Alex. Yes, looking forward to more discussions to build on the webinar that you did for us on the 6th of September. And then we'll talk a bit about what we've got coming up next because we've got an exciting lineup of further webinars taking us through to the end of the year. Absolutely. So what have we been doing since we last met? So this is a couple of pictures of Jane and I in Oxford with Amanda Wuckerook from the University of Alberta in Canada. Many of you will have seen Amanda talking at various events, such as the Fair Dealing Week special event that we ran earlier this year. And we'll know that we've been working with Amanda and her colleagues, Celine, on a copyright anxiety study. So that's what we were doing in Oxford, staying at the marvellous Crossroads College. We were just swanking about, really, weren't we? It was lovely. That's what you do with when you were in Oxford, I've concluded. I was doing my best to crack the whip and get us to get on with as much work as possible. We did actually make a lot of progress. The plan is we're still analysing the data from that. It's been really interesting. We've had over 500 responses from HE professionals across the UK and Canada. And it's a continuation of work we've been doing Focus Groups, loads of really great data. Thank you to anybody who joined us, one of those totally anonymous Focus Groups, but we will be really busy this coming term, analysing this data. And we had a really nice time at Christchurch College, I have to say, currently it's my favourite of all the Oxford colleges, but I'm very happy to go on a tour, Chris, to check out all of them so I can do a proper empirical piece of work to establish which one I think... I'm sure that you would want to... ...rank the highest, which one has the best breakfast, I mean, dining room, it does feature in Harry Potter. It is pretty awesome. We were going to put a picture in one of the dining rooms. But the thing I would say, just before we move on, is to say a huge thanks to my colleagues at the Bodleian Library staff development team. So they were actually putting on a conference that week that coincided with when we were there. So we actually presented the Bodleian staff development conference on initial findings, which was a really useful session to be part of, and actually focused on actually making us do some work, because we had to show something at the end of the... And that goes to me, that that would make a great topic for webinar in January, maybe. Well, we will definitely be doing a webinar on copyright anxiety when it happens and in what format we shall see. OK, so this is a reminder that we have the archive of all the webinars that we've done on the blog and on the YouTube playlist that's part of the Association for Learning Technology YouTube channel. So you can... I'm just going to put those links into the chat for you. Yes, yes, yes. There we go. And we now have... Copyright... Copyright news. There we go. Over to you. To spare everybody's ears, actually, Chris, I think I might have that roadcast, a thing that you want. OK, so copyright waffle, our podcast, we released another episode. We've still got a few more in the can that we're trying to get edited, but we managed to finally clean up the audio on the interview we did with Chris's ultimate boss, Richard Ovedon, who's the head of the Bodleian. I think he's called Bodley's Librarian. And we had a fantastic chat with him last summer. Actually, the reason this took quite as long as it did was not just because we've been busy doing other things, swanking around Oxford Colleges and stuff, it was actually because there was quite a serious issue with the audio. They were putting up scaffolding, which they were taking down, actually, during our visit. So I enjoyed the sound of scaffolding going up and down. Oh, you've just put the link in, I was about to do that. But we've managed to clean it up, so you don't have to listen to the sound of scaffolding being erected during this webinar. And you can hear Richard talking about burning the books, his really great book that was serialized on Radio 4. He talks about his career in libraries. He talks about the people that influenced him. He talks about copyright and the way that it's sort of impacting on our ability to preserve digital heritage. So really great discussion we had, wasn't it, Chris? It was really great. And something I think we could actually return to in our conversation with Alex. There was a point in that cleanup where we did have a go at using AI tool. Let's come back to that because it was an interesting experience, wasn't it? It was. Okay, so the next news item in the final one is that there is a webinar taking place. Spark Europe have concluded this study into rights retention policies. It's been funded by Knowledge Rights 21 and that's going to be on the 19th of October. Rights retention be the next mover, how to implement a policy supporting rights retention that they've been licensing in your institution. A very, very hot topic in universities at the moment. Something that I am right in the middle of as well and something I'm going to need to go and finish off the paper about as soon as we have finished. Sorry, was that a pun that you're right in the middle of it? It wasn't. An unintentional pun. I think it's really hilarious. You're naturally... Unintended dad jokes, it's just how it goes. Right, I think that is 10 minutes past. We have now reached a point where we can move on to the main topic of conversation. Absolutely, the star of the show today. Thank you, Alex Fenlan from University of Birmingham for coming back to talk to us again. And we are hoping that many of you were listening to the webinar on the 8th of September that we did about copyright issues to be considered when we're thinking about artificial intelligence. What we'd really like to do, I think, is... Because I think lots of you were there. Lots of you have listened to the recording. But, Alex, good morning. I think I'd just like to say hello. And maybe you could just start us off by in no more than one minute, giving us a quick rundown on what was it that you talked about on the 8th of September? Or what were the highlights for you? Morning, everybody. Thank you. Morning. Thank you, Chris and Jane. We talked about some of the issues that are coming to light in Birmingham and some of the conversations that we're having around teaching, learning, assessment and the research using generative AI tools, GAI tools. The fact that lots of people seem to be experimenting, seem to be playing with these tools and using them quite widely. And my fear that conversations around data protection and copyright and licensing have been overlooked. The horse has already bolted in some of those discussions while the focus has very much been on how do we use these tools rather than if could or should we use them? I think there's probably lots of similar conversations going on up and down the country. So, really, this webinar is probably a good idea to get insight from colleagues across the country. Share some of their practice, share some of those burning topics and some of those conversations. I'm really looking forward to the discussion today. Yeah, great. No, thank you, Alex. No, we're really looking forward to it as well. I think loads of things that sort of spring to my mind. We've got a list of questions, haven't we? We do, yeah, yeah. I think we picked some of these out of the chat from the webinar we did last month. But I think, do you want to flash those up on the screen? Yeah, yeah, I'm just getting them up. So we've got some questions going. Just before we do get into that, we also just wanted to remind everybody that you and Stephen Penton at Citi had created this video as well. So that is the guide that you've created, that if people haven't looked at it, we would strongly recommend you do because it covers the things that we discussed. Yeah, I mean, it's going live to students this term, actually, that video. So just to say it is aimed at students, but it's going also in a suite of resources for academic staff. And I've just popped that into the chat. If you haven't had a chance to look at that, hopefully it should work. Yes, I see Helen said she's obviously been able to get access to it. She said it's good and informative. Thank you. I did have to do something with the permissions to just make sure because it's on our media space platform, but hopefully I've made it on open access. And I did make the point as well to my colleagues chairing our working group on AI that it really should be something where we're operating and sharing resources. I spotted yesterday, and I haven't yet seen this go live, but I spotted them on Twitter, Stroke X, which I sort of am trying to get myself off, wean myself off it, but can't find anything else. But I spotted that there is a MOOC that UCL are developing all about generative AI. And so I think we'll definitely be sharing that soon as I've seen that when that's been launched. I think it's coming really soon though. Yeah. Christina, yes, there was an issue with that video a few days ago. Hopefully you should be able to see it now. Yeah. So should we pop up the list of questions? Yeah, that we... Yeah. So the plan with this is these were the things that we noted down last time and then looked back at the questions that were in the chat and came up with it's kind of six questions that kind of probably fit sort of a number of those areas that Alex covered. What we intend to do is put you into breakout rooms so that you can discuss it when you have kind of heard us have an initial conversation about them. So Alex, what we're going to do is just run through some of these and get your initial thoughts about it. Make sure that we fully explained what we're kind of getting at with these questions. So Jane, if you can also put these areas for discussion in the chat, that should allow people to access it when they do go out into the breakout rooms. Hopefully you'll still get access to the areas that we're talking about. So if I may start, I would like to pick on the first two points. One, knowing how to pitch the level of detail that you provide about copyright and generative AI as a copyright advisor, which I think goes part and parcel with how to become part of a coordinated institutional response. So what struck me in hearing Alex, you're very helpful setting out unpacking of all the things around the copyright implications. We know that there's a lot of complexity behind it and a lot of uncertainty because copyright jurisdictions are different in different countries. The law works slightly differently and we are in a period of flux where things are changing all the time. So it's about how do you contribute to something which is helpful and meaningful but still an accurate reflection of the situation. So Alex, I mean, what were your thoughts on that aspect of this? I mean, it's not something where there is one single answer, is there? It's just something to be aware of. Yeah, absolutely. It's about some of those core things that we as copyright advisors have been doing for a number of years. What we try to do is part of our everyday activity. So trying to tap into those networks where the communities within our institutions are having these conversations, trying to engage using the language that that community is familiar with and trying to not be too technical and approach it on a level that is appropriate for that particular audience. So our institutions are a hefty, the Higher Education Futures Institute, the department that are doing the CPB for our teaching staff, they're heavily involved in the conversations around the policy for plagiarism and academic practice and academic integrity and all of that. Luckily enough, they've just became part of an extended library division, library and learning resources division that we're now called. We were able to build those bridges quite naturally, have those conversations and it fitted into the joining up of those two previously distinct service areas. On the research side of things, I've been involved with our Institute of Data Science for a number of years and actually having those conversations again helps us tap into the discussion in the dialogue so that we can actually make sure that both communities are talking to each other, which will happen in the future, hopefully and join some of those dots together and again, using the language that's appropriate. But this is something that we've been talking about for a number of years on this forum and elsewhere as well, I think. Brilliant, I mean, that's, yeah. And from my perspective, I've just been asked to join a special interest group at Oxford where colleagues from research services and IT services are getting together to talk about these things. I'm also aware that there is research being undertaken at the moment to understand where copyright fits into that. So I think when we come to putting people into the breakout rooms, it'll be useful if it may be something that people want to reflect on as to what their own experiences are, things that they've found that have worked for them, challenges that they have, that we can all consider together. I found it quite interesting, actually, that our task and finish group, as it was called, not working group on generative AIT, actually unprompted said, oh, we want a video about copyright and AI. It's being shared by someone in our business school who I think is quite savvy about kind of copyright and IP issues. But it actually, you know, it came, it sort of, it came from the group. It didn't come from, certainly I didn't suggest it, it was on the working group, but I didn't suggest it. And then I said, yeah, and, you know, I brought Stephen in to help me make that video because obviously he's our copyright advisor. But I think it's really interesting what focus you get. You know, there was a lot of talk about academic misconduct mainly and that was actually something recently they started asking, could we edit the video because they were concerned about something I said in it that they thought made it sound like, I don't know, something to do with academic misconduct and copyright. And I think it was a bit of a misunderstanding, but we had to make a couple of changes to it because I think a lot of people are really murky about how copyright does relate to plagiarism, even academic staff. Absolutely. So should we move on to the next? I was gonna say the other topic on there about how to site work created by Generative AI. I mean, that was actually something Stephen and I looked into when we were making our video because we were very clear saying to students about obviously not to be using Generative AI for assignments unless they've been given breast permission to use it. But obviously when they do use it, how do they acknowledge which tool they used and how to do that? And we found there are some conventions about that. Actually, we sort of suggested some ways in our video that students can do it. But some of the, gosh, my brain, I don't teach cited in reference anymore, citation styles. Sorry, that's the term I'm looking for. They do actually include some guidance on how to acknowledge Generative AI. But I think the more worrying question is something that we talked about, Chris, when we were at the old conference, I think a couple of people were mentioning this to us about this pushback about openness and saying, if you put Creative Commons licenses on content and you share things openly, then it's all wide open to being harvested by Generative AI tools and used in ways that you wouldn't want or you wouldn't have anticipated. And so that's why we should lock everything down. I think it's a really interesting topic to talk about. Yeah, Alex, I mean, what's your, because you mentioned this around openness, flexibility, we're talking about things, particularly the text and data mining exception. And text and data mining is key to developing these large language models and the machine learning that is at the heart of this Generative AI. What's your take on where we are at the moment? Yeah, I think it's a real concern, but it's about citation and acknowledgement, right? And whether it's impossible for reasons of impracticality to cite, thinking about the nature of some of the wordings and some of the exceptions in the legislation and how LLMs don't cite their sources. And whether that is because it's impossible or whether it's impractical and therefore not necessary, when you look at Craig Commons, all of them have the requirement to cite the owner, the originator, the author, whatever that may be, and indeed include a link to the license as well. So while the exception might be fine because of impossibility and practicality, the issue around licensing might be a problem. We know that the exceptions have a no-contract override as well, so that balance to be struck, I think. One of the questions that has come to my mind recently is around digitization of historical archives. And if they're out of copyright fine, if they are in copyright works, if they're orphan works, if they're subject to 2039, for example, and we're thinking about licensing them openly, then how do we think about that? How do we feel about these LLMs ingesting this content and reproducing it and resurfacing it, repurposing it for non-commercial and absolutely for commercial purposes in a world where increasingly there are subscription models and revenue being generated off the back of some of these tools as well. So it's really interesting and nuanced conversation and surprise, surprise probably needs to be based on risk management and an institutional approach to open this at a strategic level, policy level probably. Yeah, because what this is coming back to at that middle point there about openness is, openness means being truly open. Openness means not trying to apply controls to things on the basis of sort of copyright and licensing reasoning. And so we're certainly, and I've just recently joined lack of the Library and Archives Copyright Alliance and talking with colleagues about sort of broader policy position. I mean, what we're seeing from, I think within education and research institutions is yes, there are concerns around these technologies and the transparency and about what's happening, what's going into those machines, what's coming out of them. But the copyright is not the appropriate tool to regulate that because if you, your copyright solution to being worried about what goes in and what comes out is that you effectively cut off all of the beneficial uses that we'd want to see of AI of using things for research purposes. But again, I think it's a really interesting point that you make there, Alex, about our institutions, we are generating large amounts of this content when we do these projects to digitize and make available and our approach has been towards openness. So it's those sort of big high level strategic conversations that we need to be taking place. Yeah. I think so. There's an issue around, sorry, sorry to go on. No, no, no, you continue. Carry on, I was just kind of making positive noises. Okay, thank you. I was going to talk about, going to mention how we as archives, galleries, museums, when we digitize our collections, when we digitize public domain works out of copyright works and what do we do around the copyright status of those works? Do we, as a policy decision, decide that actually we want to claim copyright in those facsimile plays in those digital surrogates, whatever we want to call them. And we then stick a CC license on them, regardless of how open or closed that license is. There's a question mark there around protecting the public domain and technical skill versus originality in what copyright is. So almost those fundamental questions at a certain level. And again, they are policy decisions that impact openness. And I agree, Chris, if we want to be open, then part of being open is actually being less concerned about downstream uses and being comfortable with that if that's the decision that the institution wants to take and that's policy strategy level, I suppose. Brilliant, thank you. If we just move to the next question, which I think is actually partly what Agnes' comment that she's put in chat is addressing as well, is about if there's infringement using AI tools, whose responsibility is it? Is it actually on the end user or is it something where it's the responsibility of the company that's created the tool? And I think with many of these tools, they're issuing license terms and conditions that say this is, if you use this tool, this is what you've signed up to. And so Agnes, you're talking about the co-pilot. What do they call it? Is it a tool, a service? It's being offered anyway by Microsoft. I've just heard that we're not going to be switching that on at my university any time soon. I think we're going to be testing it out and seeing what that does. But I think it's a, where does the responsibility lie and can these companies kind of wipe their hands of the matter? If they say, well, nothing to do with us, it's all down to the end user because you're using the tool. So, yeah. It's a really interesting question. I mean, looking at those terms that Agnes has posted in there, is that a warranty? Is that an indemnity? So will they indemnify the user, the institution against any infringement of third part by a third party based on their use of co-pilot? I mean, that's really interesting and really significant as well, I think. I can imagine our institutional legal departments having a field day and also having kittens when they see, have sight of that, if they indeed ever get sight of those agreements that are pushed through. And of course, co-pilot at and GitHub is subject to legal action as well, right? So it's going to be interesting to see how that develops over the coming, I was going to say weeks then, but it's going to be months and years, isn't it? Yeah, interesting and thanks for that, Agnes. It's really important. Did everybody see the post from, I put in the chat, sorry, on a list copy-seek earlier today about JISC's National Centre for AI and the... Not everyone when I've seen it, Alex. I was going to say, I can pop that in the chat, yeah. Yeah, I think that's worth. Yeah, thank you. They did a review of some of the terms and conditions of some of the tools, talking about who's able to access the tool but also about how they reuse user input. And that's really important, really useful to see somebody actually summarizing that and looking at some of the terms and conditions. And that's an eye-opener, right? How disparate and how varied the approach is between the different tools. So yeah, if you haven't looked at that, because that's going to be quite useful and quite informative, especially when we are advising our colleagues as to some of the issues around use of some of these tools going forward. Yeah, I'm just... I'm still waiting, sorry. I will be putting that in the chat. I'm just waiting for a few things to load. Everything's grinding least low on my computer at the moment. So, but we're getting there. And then the final question is just about whether the kind of has the horse bolted when it comes to the, you know, in terms of use and ownership of IP. And, you know, I mean, people are out there using all these tools, aren't they? So, you know, it's only kind of now that we're starting to say, oh, hang on a minute. What about X, Y and Z? I think, I mean, I think that goes across the board for all the kind of ethical issues associated with the use of AI that we may well, you know, we're kind of plunging into this brave new world. And I think lots of things will probably come to light down the line that hadn't been thought of. Some of which have already been played out in various sci-fi films. Yeah, indeed. I think that's really interesting. I think, you know, since the summer, probably Easter really, there's been a real focus on the learning side about how we use these tools, how we upskill our academic colleagues, students in using some of these tools. And I think slowly we're starting to see that horse come back in. Maybe it's approaching dinner time or maybe it wants a safe place to rest its head, come back to sleep. You know, coming back into the stable where we're saying, right, okay, there's been this hype around this activity. We really do need to start thinking about some of these things. And I think this post from the National Center for AI in tertiary education is gonna be really interesting to actually bring that to focus in some areas because I know I've struggled to gain traction previously. And this is gonna help us in those conversations, I think. Yeah, I think this is a really interesting point is because we know that people, click, agree, all the time, we all do it, all the time, and we're just going through it. But I think with this particular topic, because the nature of AI, generative AI, and that sense of the fear, the concern, or the amazing opportunity that we have, I think people are kind of having to stop and think about what it is that they're actually signing up to. If it's something that just is using, these technologies a bit of around using AI and algorithms to sort of push things at you that seem to know what you're like. But now that we're actually seeing something which is getting at the heart of what it is to be creative and to be a human being, I think that's where, as you're saying, you could bring the horse back. I mean, it is a necessary thing to do. We can't just simply say, oh, these tools are out there and you can't stop people using them. I think people want to know what the right thing to do is or to be informed enough to be able to make their choice about which tools they use. Yeah, and I'm a research skills team who recently developed a webinar, a workshop that they run with the business school. And when they were approached for that workshop, it was very much a focus on what are the tools that are out there, how can we use them? But research skills, rightly in my view, approached it to say, these are some of the issues, these are some of the literacies that we need to address in order to be able to... Right, in ethics and a practical basis, the library simply is in that position. And there's nobody else to do that work. But also we can't just simply wear the one single copyright hat when we know that what people want is a range of expert positions and guidance into trying to develop positions and guidance. There is loads in the chat. I don't know, we've only got a couple of minutes left. I mean, I was in a really interesting breakout group. We were just talking about maybe staffed, sometimes feeling a bit nervous, a bit unconfident about being asked to be involved in advising on AI. But my experience has been that actually there's many of us who sat around the table trying to draw up this guidance that haven't got the answers. So that's not a reason not to get involved because I think if you feel like I'm not an expert in this, then I think there'd be very, very few people in these working groups because it's just something where we've all got something that we can contribute. So I just urge people to not shy away from trying to get involved in this and to just be really honest about where your expertise is as well. Because it's certainly, when you think you don't know that much about it, I sort of found that I probably do know more than quite a lot of people. But then also, I've got colleagues that clearly have been working on AI and doing research on it for years. And so I'm gonna lean on them as well. So, yeah. Agnes has just asked about whether we should use different types of citation depending on if they're created in the UK or whether it's... Yeah, I'm not sure what the benefit of that would be. It depends on what law applies or whether there is a difference in what actually is going to be created. I'm not sure that there is a huge value in that unless I'm missing something. I think given where we are with the time, I think we need to draw the conversation to a close. I mean, I certainly found that very useful. I hope everybody else did. Alex, any final words from you before we kind of wrap up this part of the session? I mean, Helen's comment there made me smile. Students have always been cheated, but maybe it makes it more visible or less visible, perhaps. I mean, it's never a true or word spoken, I suppose, but yeah. I'd agree. I'd agree, yeah. And to echo the comments you made about we might not think we have expertise, but absolutely we do. Absolutely engage, have the conversations, ask the questions because they're important. Yeah, brilliant. What a great thank you, Alex. Thanks for the end of that. Yeah, thanks so much for coming back. I'm sure we're going to pick this up in the future. I can see the copyright and AI discussions are going to run and run. So, you know, watch this space, obviously share resources that you've got everyone. We've got a couple of upcoming webinars and we've got a couple of webinars we've got a couple of upcoming webinars before the end of the year. So 10th of November, we're going to be talking about open textbooks. We've got David Beals from Brunel. And David spoke at Icebox, didn't he? And he was a brilliant presentation. He only had a lightning talk. So for even people that were there, he's got a lot to say, very engaging and clear. And I think he's got a number of colleagues joining him as well. Yeah, so that's going to be a fantastic one, definitely. Yeah. And then we've got a really exciting thing at the end of the year, haven't we? We've enlisted our correspondent from IFLA, Matt's voice, to help us run what we call in the Big Fat Copyright Quiz of the Year. It is. It's our Christmas special. So we're getting ready for that. That's going to be brilliant. We hope to see you there. And then we do... Are we putting people in teams? Is that how it's working? We haven't quite worked out the format, but we'll have something that will be excellent. I'm not sure. Will it top the blockbusters that we did the other year? I don't know. We'll have to wait and see. So that concludes today's webinar. For those of you that loves things. One last thing, though. Slightly beyond the bitter end is... Do you even explain what this is all about? Well, yeah, I can explain it. You know, many people know Chris love showing off, and he had a great chance to show off last month a gig that it was in celebration of George Harrison and all his music. And you were actually performing with some friends and with a group that was put together that you decided to call the Wondering Elder Berries. Wondering Elder Berries, yeah. Was it a kind of, was it a parody, a pastiche, or was it just a cover version of a Travelling Willbury song? I was at a dog's dinner. Shall I play it? Yeah, I'm going to... Shall I stop the recording, though? You don't want this on the recording? Yeah, I think we should stop the recording now and say thank you to everyone that joined us. Anyone that wants to waste another 38 seconds. Anyone that wants to hear through this impersonation of Roy Orbison, then it will get a take-down note. It will get a take-down notice on YouTube, so that's why I'm stopping the recording now. So we'll see everybody next time, next month, but yes, if you want to stay and listen to him, then I think...