 Okay, let's go, and that makes you happy. Good morning, everybody. Good morning! So my name is Chris Morrison. My name's Jane Secker. And we are the co-chairs of the Association for Learning Technologies, Copyright and Online Learning Special Interest Group. And this is our 58th webinar. 58th webinar that we've run since March 2020. About copyright and online learning at a time of uncertainty. And just general great conversations about copyright and keeping everybody up to date. So we're delighted to be back here again to have many regulars joining us. And to have our guest with us. And a great guest. He's going to be talking about copyright and artificial intelligence. So we're really looking forward to it. We'll come on to that in a moment. So this is just the overview of what we've got. We've got quite a few bits of copyright news, some news items, many events coming up, many exciting events. But as you say, we have Margarita Vindesh here joining us from Zurich to talk about copyright, AI and artistic works. A very hot topic. Absolutely. And we're going to dig into some of that. Yep. So what's happened since we've last met? We've got a couple of points here. These are, I think these are health and well-being related, aren't they? Yeah, we're definitely. Yeah, it's not January anymore, but it's February. So, but we still try to keep ourselves happy and healthy. And we should say, this webinar is not sponsored. So these products that we've been, we're entirely our choice, but I mean, many people are horrified by my choice of footwear here, which are the vibrant five fingers in a way to... I think we need some comments appearing in the chat that of what people think actually. I think it's one of those things. Don't let them see them in real life. If you know about, if you know, you know. Right. If you're not in the club, then you might find them horrific. But actually they are up in it, right? Right. Excellent. But you've been, you've also been doing something to improve your... I have, yeah, yeah. I've discovered, I'm telling everybody about the benefits of taking turmeric or turmeric or whatever you want to call it. So this is something that I started taking about three weeks ago. I felt like a new woman. You do, indeed. Yes. So just recommending a couple of things there. And we've got some... Maybe not the shoes. Lisa says, better foot shoes all the way. She's in the know. Okay. Okay, right. So we just to say, we have an archive of all the previous webinars and blogs on the copyright literacy.org website also on Alts YouTube channel. Yep. And we've got some links going in there. Thank you very much, Greg. Oh, yes, thank you. And here we go. It's time for... Copyright News. Copyright News. So news item number one, the UK government has axed its plans to broaden the text and data mining exceptions. So this was news we brought you a while back that there were plans to broaden this out to allow commercial text and data mining under copyright exceptions. And they have chosen not to do that. Interesting process. They had a consultation. Yes. And then they just made their decision and then they reversed their decision. Yes. So I mean, that's a whole other conversation really about the process of lawmaking, advocacy and lobbying. But certainly I think something that we will likely be returning to in our conversation later on today because text and data mining computational analysis is so central fundamental to the area of artificial intelligence. Yeah. But that's hot off the press. This is also hot off the press. It is. It's up on the copyright literacy blog. It's the annual report of the activities that we've been doing under the all species of the Alts Copyright and Online Earnings Special Interest Group for 2022. So if you want to know what we've been up to in addition to running these webinars, which obviously feature, then that's the place to go. Have a check out our annual report. Absolutely. The next news item is an event, the Knowledge Rights 21 program. We've been featuring quite a lot of the work they've been doing. They are holding an event on flexible copyright exceptions next week. Yeah, Thursday. And that is on the 13th, which is Monday. It is, yeah. So that, again, is going to be a really useful and interesting discussion there about flexible copyright exceptions in different jurisdictions around the world. Now, copyright. I think we've got Emily Hudson talking, haven't they, as well? I think. I believe so. Today, hello, Emily. So looking forward to that one. Next news story. Yeah, next up. So last year, for the first time in the UK, we ran a number of events during what's called Fair Dealing or Fair Use Week. It started in the US. And we're really excited to be running an event in conjunction with the Institute of Advanced Legal Studies. It's going to be on the 20th of February. And we've got a fantastic speaker, Amanda Wackerook, who is the Scholarly Communications University of Alberta. And we're also going to be joined by Carl Cagel, Courtney, who will be coming on. Ile himself. Yeah. So that is now available for booking. We've put a link in on the ILAS website. If you want to join it, it's a webinar, so open to all. And it'll be five o'clock in the evening. So just the kind of thing you want to, you know, unwind on a Monday. Absolutely. Amanda's a fantastic speaker. Some of you will have the work that we've been doing with her. We're looking forward to this session. But that's not the only Fair Dealing week session coming up. So the Scurl Copyright Group, it's Copyright and Legal Group. So that's the Scottish, it reminds us what Scurl stands for. It's your university and research libraries group, isn't it? And I've put together yet again, another fantastic program here with some really great speakers. We've got Zoe Crochita, who spoke to us about filter, upload filters and the relationship to the digital single market. We've got Emily Hudson talking about the work she's doing, looking into scholarly monographs and open access. Open access, absolutely. So really looking forward to that one. Yeah, yeah. So lots of things to put in your diary if you want to go along. And another event. Another event. So this is one. What is Oxfords? The Oxford Festival of Open Scholarship. This is one of the key events that the team I'm part of, the Open Scholarship Support Team at the University of Oxford based in the Bodleian Libraries does a two week event where we primarily promote things to Oxford researchers and students and everyone who works there and is part of the Oxford family. But this is an external event that will be happening in real life, in person at the Western Library at the Bodleian Libraries. But the reason why it's quite short notice is because we've actually been working on the feasibility of running it as a hybrid event where people can remote in. And I'm delighted to say that we are running it as a remote event. Haven't you got the bookings up there? So what Greg has posted into the chat is an expression of wish interest or expression of interest for and we'll send out the invitations. But we'll also do it on all the mailing lists, copy, copy, so. But it's a discussion about where the copyright form is necessary to achieve an open access future. We have Professor John Wilinski from Stanford talking about that. I saw an Estefe who some of you will have heard on this webinar. She's a friend of this webinar and did a great copyright waffle. She did. We've got Chris Banks from Imperial and Andy Redmond from OUP. How do you go in to perform the song as well? The song about Asun coming to the UK. Yeah. I think you should be there taking your guitar. We will see. We'll see whether that fits into the Bodleian context or not. Yeah, I think that's Western Library that we go down well. Bit more rock and roll. I'll make those notes. But that one's really exciting. Excellent. At the end of all the exciting things, is it? No, it's not. It's not. We're going to be speaking at the end of March at the University of London's RIDE conference, which is Research in Distance Education. So it's run by their Centre for Online and Distance Education. And there is going to be a panel at the end of the Tuesday, which I think is the 28th of March, where we're going to be looking at open education and kind of how we can kind of turn that into sort of from theory and to practice. And there'll be a line up of speakers that Chris and I are just two of the members of. But the conference programme is now up on the website and it's available for bookings. And I believe it's being run as a hybrid event as well. I believe there isn't a registration fee if anyone's in London and they want to come along to that conference. Yeah. But even that's not the end of the exciting news, is it? If it couldn't get anywhere, it's like, at this point, I needed party poppers. Oh, right. OK, we don't have any party poppers. Can you play the theme tune? Many... The ice pops theme tune. I think we might need to get this together. We don't actually have ice pops theme tune, do we? Don't know. But the International Copyright Literacy Event with playful opportunities for practitioners and scholars. Many people here are aware of it. It is back. It's coming back. It's coming back with a vengeance. It's going to be brilliant. It's happening in Glasgow. At the end of July, we are partnering with Create, the research centre for copyright and the creative economy, whose work is... It's going to be a bit like a festival, isn't it? We're thinking about it as a festival... Festival of Copyright Literacy. It's like a camping, on-site camping as well. Glasgow in July? Yeah. I mean, maybe not in the middle of the city. OK, OK. I'm just thinking of, you know, the pyramid stage and all, you know, it's sort of, if you haven't got a ticket for Glastonbury, this is the next best thing. I would say it probably is, yeah. I mean, we've got one keynote confirmed, who is Professor Nick Whitton from the University of Durham. Nick is one of the leading experts in playful learning. So we're returning to the concept of playfulness and games and gaming industry. We know Create have a lot ties in Glasgow and Scotland. There's an awful lot of creative stuff in the game industry space. So we're delighted to be doing that. More details coming, but please put the all we want call for contributions is open. We want your contributions to share your creative ideas about copyright education. The booking is not yet quite open. The book is not quite open. Save the date and think about if you want to come and speak at the event and also kind of watch this space for more. We will get the bookings up in the next couple of weeks. So before we move on, I just want to acknowledge James, James Bennett, your comment about ride and shoe gazing. Yes, noted. I don't think there'll be lots of sort of swirly, psychedelic indie guitar noise at that particular event. But maybe we could we could introduce it. I don't know. Another festival. Another festival. A lot of time. Let's turn all academic conferences into festivals. Now, it's quarter past. It is. It's time to move on. Yes. And we are absolutely delighted to be welcoming guest today, Margarita Vindesh, who is a PhD candidate at the Centre for Law and Economics at ETH Zurich. We met Margarita actually when we were in. Yeah, great. At the Advanced Research Centre. We had a really great conversation with her. Now, Margarita's area of specialism is artificial intelligence and copyright, particularly as it relates to art and artists. So we won't say too much more to steal Margarita's thunder. We we've spoken to her about the links between copyright and online learning, which is what we're covering and the artwork and an art world that the Margarita looks at. So there are some links there, but we're really going to hand over to Margarita to share with us what she's been working on and give us some insight into what is clearly a very hot topic. A lot of people talking a lot of stuff about AI and we're really glad to have her putting it into some kind of sense and make it happen. Absolutely, yeah. So Margarita, are you are you receiving us? Can we hear you? Yes, I'm here. Excellent, right. I'm going to going to stop the sharing house lights. We will get your slides up. The floor is yours. So also, so thank you, Jane and Chris, first for the very nice introduction and hello to everyone. I'm very happy to be here and to talk a bit about the pressing question regarding copyright law and AI and also about my background and my research in this topic as a PhD candidate. And to give you a bit of a background for myself, so I have a classical legal background where I was already like fascinated by copyright law and intellectual property law in general. But what many of the people don't know is that I also have a creative background. So I went to a fashion school before my legal studies. I come from a very artistic family. And so now the PhD allows me to combine the creative part with the legal aspect and doing research on that and also to connect the different stakeholders, such as artists, AI developers, and legal scholars as myself, which is amazing. So I love what I do. And everyone hears about AI in different contexts are almost every day in the news and how it is disrupting the creative industries and the work of artists, but also what positive impacts AI has on their work and the future of creation. And then there are also some questions regarding copyright infringement and the future of creators in general. So it is a very exciting topic to research on and to dive deeper. And for today's webinar, we will first briefly get into the basics of copyright law, especially in the digital age, and then dive deeper into copyright law in the context of AI and discuss the questions of copyright authorship and copyright infringement, and then also what are the possible remedies and how to continue in the future. To get everyone in the same page, copyright law provides creators of original work with a bundle of exclusive rights, be it for literal, musical, artistic works that are fixed in any tangible medium. And in most countries, the duration is the life of the author plus 70 years. And it is limited to the expression of ideas and facts, but not the ideas itself or functions or facts are not protected and default automatically into the public domain. And next to the economic rights, as copyrights there are the moral rights, which give the creators also personal rights, like the right to be attributed or the right to object to any alteration of their original work. And we already saw in the last decade that technology significantly shaped the way creative works are produced, distributed, and used. We have seen an increase in created works that are also then made available to consumers, and we haven't seen a decrease of costs such as in production, distribution, and advertising. And already back then or until now, one of the most pressing question was how do we identify and prosecute copyright infringement online? And this got now even more difficult in the context of AI, where the first question is, can the output of generative AI models, can it be copyrighted? And if yes, who owns the copyrights or who's the author of the output? And also regarding the question of copyright infringement, is there any infringement happening? Do copyright owners, where their work is being used to train the AI, have any legal claims over the model or over the content that it creates? And based on those questions, we can then also look at, how should we deal with the consequences of AI in the creative industries? What kind of legal constraints or restraints should or could be put in place? And also, how could the relationship between AI developers and copyright owners but also the users of the AI could look like? And what makes it even more complex is there are different stakeholders involved with different interests. So we have the creators whose work is being used to train the model. Then we have the AI developers who develop the model. Often it is an AI company. And then we have the user of the AI, which are also often creators that use AI as a mere tool. And then we have society as a whole. And to tackle the first question I mentioned regarding copyright authorship, we so far have human authorship as a guiding principle. And there are three examples you see in the US context of copyright law, where you might remember the monkey selfie case where a photographer gave a camera to a monkey and the monkey took all the pictures in the jungle. And there the Court of Appeal of the Ninth Circuit said that those pictures are not copyrightable because there is no human as an author. And also, in case of registering a work, the US Copyright Office so far said in one case that there is no registration possible for a work that is created with an AI. In the case of Tala and in the case of Kastanova, which you see on the right side, there has been a registration of a novel where after the registration it has been revealed that the images were generated with an AI. And so with the urge to question now, should the Copyright Office remove the registration or leave it? And so this is not clear yet, but very interesting to see how they will pursue. And we can assume that also computer-generated works have human authors so far, which us don't know who it might be. And the UK, for example, is one of the very few countries that already protect AI-generated works where the UK Copyright Act states that the author is the person by whom the arrangement necessary for the creation of the work are undertaken. However, it is not clear what are the arrangements that needs to happen, what is necessary. So this is still up for discussion. And there's also a shorter protection period for the works. So it's the life of the author plus 50 years, such as for photographs or databases, and also more rights are denied for those works. And so the general question is, what does it mean to be an author? And when should creative works be protected? And here you see two pictures of images where one is created with an AI, and one is created solely by a human being with maybe other technology or more traditional tools. But it is not clear from just looking at the pictures or the images which tools were being used. And if AI was involved or not. But in both cases, and I can reveal the question of the AI-generated work afterwards. But in most cases, also when work is generated within AI, a lot of effort goes into creating also the output, not only the model, but then also working with the AI to generate something. And you have many creative decisions along the way which are important for the originality of the work so that it is protected by copyright. And there also my current research project focuses on where I look at the question of whether the attachment between artists and the work depend on the involvement of AI. And I focus on text-to-image models in the context of visual arts. And in order to measure relevant factors or measure the input by human being, there are several relevant factors, such as the initiative to create the work, the investment that goes into creating the work, and then also the extent to which the arrangement shape the work, the form and how they are responsible for the materialization of the work. And also one factor is the proximity to the act of the final creation, which is also interesting because often, so copyright law has different theories underlying such as the utilitarian approach, which is more incentive-based, but then also in Europe you have the approach of the labor theory, which states that it is about the investment that goes into creating the work and one should own the fruits of their own labor, their intellectual labor in this case, and also the personality of the artist that is being expressed is much more relevant in continental Europe. And so there are different implications of a weaker or stronger bond to works that are created with AI compared to works that are created with other technology, but also more traditional tools like brushes on a canvas, for example. And there I'm working on an experiment at the moment to answer this question and work together with different artists, which brings us also to another important question in the context of AI and copyright law, the question of copyright infringement, where it is not clear at the moment if there is a copyright infringement, for example, when the AI is trained on databases that consists of original works that are copyrighted, but without the permission of the rights holders, and then also if there is a copyright infringement, who infringes the copyright? Is it the person or the group that puts together the database? Is it the group that uses the database or maybe even the user that uses the AI model? So it is not clear if there's an infringement and then also at what point does the infringement happen? Importantly, copyright infringement is the actual copying of the copyrighted work without permission. And at the moment, lawsuits are already flying in, so we have Getty Images suing Stability AI and Getty Images is one of the huge stock image platforms where they state that Stability AI used their work, like millions of works, to train their model, but also artists are coming together to sue different AI companies. And so it's very interesting to see how this will involve in the future because there's a lot of uncertainty and that's also why so far people are also a bit more on the safe side when it comes to suing different AI companies because it is very costly and you just don't know the outcome. And often, so for example, in the case of Getty Images, they stated that they're not interested in any financial damages or stopping the development of AI art tools, but they find it important to create like a new legal status quo because of all this uncertainty. And what is also interesting to do also for you is to use Have I Been Trained, that's a website where you can insert an image and see if it is part of the LION database, which is also the one Stability AI created or used for their model. So this is or might be also very relevant in the future and it's also fun to play already with it now or very relevant already for the creators. And to tackle the question of copyright infringement, it is also important to notice that there are exceptions to copyright. For example, in the US, you have the Fair Use Doctrine, where section 107 of the copyright act lists some of the fair use cases but then also that four factors need to be in place in order to have a fair use case as it defends to copyright infringement such as the second work is transformative. So the purpose and character of the use has somehow changed then also the nature of the copyrighted work compared to the second work. And the degree of substantial similarity is also an important factor and the effect they use upon the potential market for the value of the copyrighted work. And in the European Union, but also in Europe in general, most countries have only an exhaustive list of what is considered an exception to copyright, which includes also text and data mining, which is the key to the question more or less because AI companies or the developers or the people who create the databases they scrape the internet and get all the images, in the case of text to image models, but also all the information in general when it comes to JetGPT for example. So they scrape the internet and extract the data and this then creates the database that is used to train the model in order to create the output. And there the European Union has three or two articles related to text and data mining in the DSM directive and the copyright directive, where Article 3 for example states that text and data mining is allowed for purposes of scientific research without any permission of the rights holders and Article 4 states that it is possible for any purpose, but there needs to be an opt-out provision for rights holders and there it is now again like there is uncertainty because we see also some gray areas where AI companies, stability, AI funded a research group to undertake the model and the model training and also a small nonprofit organization which was then a lion for the database to create their data sets for training the model, which now contains about 5.8 billion images. So it's huge and importantly also most of the research that is going on in AI models and the generation of AI art for example is open source. So it is very easy also for other researchers to then generate new models and train them with the data sets that are existing. And so there's the question of balancing the rights between rights holders and also the freedom for people to innovate. And potential remedies can or are very complex at the moment to tackle because we have to consider first that there is like a complex interplay between the different stakeholders that are involved with often different interests and on top of that we have huge differences already within the or between the creative industries on a national and an international level for the music industries already really complex. And the software industry works again very different and the industry of visual arts is now very challenged in this context with AI models. And so to allow AI to develop further and to promote the innovation but also giving credit or compensation to the creators whose works make the field possible there are different solutions for example fair learning which is based on the fair use case where depending then on is it fair to use the data or not in the US with the fair use doctrine but also then in Europe with the different exceptions to copyright. But we could also think of a market-based licensing solution where for example again in the music industry after the Napster case where it was a huge case regarding copyright infringement from peer to peer services and the download of music illegally. But there's Spotify for example found a way of bringing together the rights holders of the original music then also the users of the tool and negotiating something that works for everyone. But we could also think of funds to compensate individuals whose work is used by AI companies to train their models where this is already happening a bit in the case of digital devices for example where there is a certain fee that is put on the device when you buy it that goes then to the collective society and then distributed among the creators which could also be a solution for the AI tools in the future. But there's still very open questions and it's also the question of should we regulate it now already or to what extent or should we still lean back a bit and observe what is developing and how much AI really influences the creative industry and the creators for example in case of their creation processes. So yeah a lot to look forward to and to discuss also not only today in the webinar but also in the future. And with this I thank you for your attention and looking forward to your questions now and also if you want to contact me after the webinar feel free to write me an email. Yes. Thank you Margarita, that was absolutely brilliant. Yeah, I really enjoyed it, thank you. I'll put your email there in the chat so people can grab it. So many things to pick up on so many thoughts that come and I think when anyone starts thinking about artificial intelligence, relationship to creativity and regulation of that, your head starts spinning. So I mean the work that you're doing must be fascinating and to dig into all of that. So I think I'm going to pick up on some of the questions. I think the first thing I wanted to cover though actually is you mentioned for example the United Kingdom here in the UK we have provisions around ownership of computer generated works and we've had that in our legislation for some time. And I guess we had you say it's an assumption there but there are human owners of computer generated works because it is the humans or the organisations that are behind creating that computer software the model whatever it is that's creating the work of the output of it. And what's the difference between where we were there? I mean were people fairly comfortable and confident in understanding how that worked from a legal perspective back when that computer generated works legislation was first interpreted? And what's the difference to where we are now? Is it simply that the technology is so sophisticated that it's possible to create works that are indistinguishable from human created works? So as you said it's very complex because there are different stakeholders involved and also AI is used not only by users that don't consider themselves maybe creative or an artist but it is also used by artists as a tool. And so there I think coming back to the labour theory and the investment that goes into creating a work there are differences in the advancement of technologies but I think you can still, even though if it's not distinguishable if it's like a human being or a machine that is somehow the tool or the one that moves and creates the work it is there are still many creative steps and creative decisions along the way where you can see AI as a tool which most of the artists do I think and there it should then not make a difference. What is difficult is that when you just play around with it or there has been also now a case in competition with photographs for example where the winner of the competition revealed afterwards that the work was created within AI and so there all the other artists then said okay no this is not okay this should not be allowed because every other artist put a lot of effort in creating the image with camera which is also technology which simplified the paintings for example and already back then everyone said that no we should not protect images created by cameras which is now long ago and today it's totally normal that photographs are also copyrighted and I think there even if the first prize went to an image that was created within AI it doesn't mean automatically that there was not also a human being. So I think the point you make about photography is fascinating as well because we and then we spoke to Lionel Bentley Professor Lionel Bentley on our podcast a while back and he talked about the moral panic and the so a conceptual idea of people capturing light itself and the back in the end of the 19th century. I think following on from the idea of competition and whether or not it's regarded as cheating clearly there's been a lot going on in the online learning space around using AI tools as a way of students taking a shortcut just doing a quick thing and then the computer does all the work for them. To what extent is this are you aware that this being discussed in art schools? What kind of link is there should there be between legal scholars like yourself and with an artistic interest and talking about this in a critical way within our educational institutions? I think in general it's especially now also with chat GPT for example but also in art schools with AI art generator tools it is really important to not only or not just banning those tools from the education system but to find a way to bring them into the education system and make critical thinking more important and finding also ways to integrate tools that are available not only with AI but also with digital technology in general to make people understand what does it mean to use technology as a working tool to also learn how to integrate the tools in the work where it is maybe also relevant or much more informative but I think for this especially the education system has to evolve much more to find a way that you use the tools that are available in a way where it makes sense and to also critically question the tools where do we want to use it, why do we want to use it and how do we want to use it and I also like during my or between my bachelor and my master's I worked in digital education where we provided Austrian schools with the devices but also with workshops on how they could integrate the digital technology that is available because we are surrounded by it already in the daily life so it only makes sense to use and to learn to learn how to use them in the education system where it is a safe space and where the knowledge and the connection can be created much more easily that's really interesting I think it's absolutely I mean I've tuned into a couple of talks that have been happening this week in higher education of our chat GPT I've got the link to share from the Oxford teaching and learning I'll find that yeah but I think that's absolutely we've kind of got to accept that these tools exist and teach people how to use them haven't we for kind of learning but also to reward the kind of the critical thinking which is what we want our students to do there's a question from Deborah Fern so we've actually got lots of questions in the chat so we're going to try and get to some of them but I'm kind of going back one of the first questions Deborah asked was about can you still hear me because you're okay they're gone but Deborah you can ask the questions just like this if you want to yeah I can't hear Jane at the moment but I guess my yeah my question was really about how relevant are the terms and conditions of the providers of the AI software when it comes to who owns copyright between the software provider and the user of that software I guess for context we are running on images and research competition at the university and we're really kind of trying we're trying to determine whether the creators who are submitting their images and telling us upfront they're created by AI whether we should be attributing them and they own them or we should be concerned about what rights the the model creators have and we have looked at various terms and conditions in that respect so so far every AI company is doing more or less what they want and like they try to be as safe as possible with incorporating in their terms of conditions often that the user has all the right to to use the images that were created but some rights also then remain with the AI company but at the moment it is just not clear so I would say that what is in the what is stated in the terms and conditions of each company for example stable diffusion where it might change in the future depending on those lawsuits for example but at the moment it is just unclear so take them into account or not but yeah I think I think most of the examples we've seen there there the user is the rights are assigned to the user in what their output is I think the only example I saw really that was different was the shutter stock AI often it's it is only licensed to the user so ownership or the AI company then states that they have ownership licenses all the rights to to the user of the tool but this is the question I'm also very interested at the moment to research because it is not clear who is the person necessary for all the arrangements made thank you Margaret and thanks Deborah for stepping in there that was very helpful sorry we lost a bit of connectivity here we're not quite sure what happened if it happens again we'll switch to a different device yes yeah yeah more questions I think they might be in the in the chat as well aren't there do you want to pick up the next one Chris I just a comment here from Jess about encouraging them at Manchester thinking about art and authenticity in a in a philosophical way I think that's one of the things I find we do go this is a strong ethical element and philosophical one that again makes us it's all those kind of age old questions what it is to be human and what art is and what it's about and what it's for so I imagine you spend quite a lot of time thinking about that sort of thing Margaret or do you spend your time thinking really focusing on a kind of legal and economic question yeah I think it is an interesting question and I'm also doing interviews with artists at the moment to also get some more background and there are artists who work with technology in general or more and more and to not only use it as a tool but also to showcase or like to have again like a critical view on the technology because I think that's also what art is about to some extent that we take what is happening in in the real world in society at the moment and then somehow transform it into art and so there are artists who already use technology in different ways and now this is just like another tool to create to express their ideas and they still sit 20 hours or more on that specific work to create but then there are also artists who yeah who have like a more negative view maybe on the technology and maybe also on how society is changing and evolving with the technology but also there I think art is a very good tool to showcase the issues of society and of digital technology absolutely yeah and it's just an interesting question I've spotted from from Evelyn Webster have AI companies violated the terms and conditions of social media or image hosting companies if they scrape the users uploaded content we were just having a conversation as well about music weren't we a few sort of we use a lot of music whether that's done under license or you know because I've heard of somebody who went through all the work through the Beatles to write new songs and I think related to that there's also a question from Becky about would artists have the opportunity to remove their works from the database so can we cover those at the same time to what extent is it is it a breach of the license terms and then what can you do about it if you do find something is incorporated so there comes the exception to cooperate with text data mining into place because so it is legal to scrape and to collect data in the case of research activities where you don't need the permission so if a research company or if a research group of the university scrapes and scrapes the internet collect the data trains an AI model where Dali also started as this there should not be any copyright violation but then it gets difficult if the pool is then also commercialized and being used by a company for example to make money from it and there another exception in the European Union in the copyright directive at least is that you can scrape and do text and data mining for any purpose but you need to provide an opt out possibility for right holders or you license or you get like data that is licensed where you have the license from it which is difficult to or at the moment at least it is difficult to do that on a large scale but with the opt out version this might be possible in the future then the question is also is this really enough for the right holders that where their works are being used to train the model open question open question and an area that is moving very very quickly Marguerite we've run out of time and what I'd like to say is I mean will you share with us the outcome of your research how long have you got left on your PhD studies when are you future complete I still have about two years so excellent well okay well we'll stay in touch and absolutely and that was an absolutely fascinating overview of where we are with things everybody is thinking about this and talking about it all the time so thanks for yeah thank you so much for joining us really thank you for having me yeah we hope maybe you'll come back to Glasgow who knows yeah yeah lovely thank you thank you so much right in the remaining one minute we will just finish up we've got a couple of things to share with you before and it's time to go yeah I've got lots and lots of thank you coming in in the chat I don't know if you had a chance to look at it Marguerite I think everyone found that really I'm looking at it now yeah yeah yeah lots and lots of chat so so all revealed for the one last thing too right okay so we have our next webinar is going to be a closed discussion about the CLA license and what we're looking for within the UK higher education sector because we are about to start negotiations for the license which starts out of first of august 2024 yeah we've got our first meeting next week we do so looking forward to get back into the swing of that after that we've got our becoming a copyright specialist theme coming around again we have two speakers lined up we are looking for a third if anybody is interested in giving their story about how they got dragged into the world of copyright and how what it is that keeps them within the world and to stop them running away then please let us know and then we're delighted that in May Matt Voitz from IFLA the international federation of library associations who's creating what is turning into a real blockbuster copyright news newsletter that he does for their copyright and legal matters group he's going to be sharing the highlights recent topical things things that he's really found interesting over the last few months that's going to be a fantastic and we haven't yet scheduled anything for june we haven't july is going to be busy july is going to be quite busy we'll probably take that off yeah taking it off are we we're doing ice pops no but if there are any topics thoughts yeah okay so one last thing play that jingle chris no don't play the jingle that takes too long yeah the jingle goes on forever one last thing so um i'm part of the move towards this kind of hybrid working um in my office at city i'm kind of clearing out my desk because we're looking to have a we've got a hot desk in system in place and one of the things i found in there was um some sort of six seven year olds rather stale copyright fortune cookies shall we see what they actually say on them yeah let's have a look yeah looking for more help about copyright visit copyright user dot org for advice videos and many useful resources copyright user brilliant website still around Bart Milletti yeah creative director at create and all the great people that have done that and also there is a new version of it come in there is yes yes so that's very exciting and then the other one the second one was about taking images from google may infringe copyright and it was telling you to use i think that creative common search doesn't actually exist anymore but obviously the principle still applies yes so the moral of the story is that fortune cookies might go stale but copyright does never does why would you encapsulate copyright AI and art into one snappy fortune cookie statement probably quite i'll have a bit of a think about that one yeah i tell you what just get chat gpt to do it oh that's a good yeah write me some copyright i'm sure we get it to do it this afternoon right some copyright um fortune cookie motto is to see what it comes up with shawn as pointed out it's openverse.org there's given us excellent yes thank you brilliant okay well i think we will leave things there thank you again for everyone for turning up again thanks again to margarita for such an excellent presentation and we wish you a lovely weekend yeah bye everyone