Warfare of Art & Law Podcast
Warfare of Art and Law Podcast sparks conversation about the intriguing – and sometimes infuriating – stories that arise in the worlds of art and law with artist and attorney Stephanie Drawdy.
Warfare of Art & Law Podcast
AI & IP Panel Discussion: A Global Perspective Part II
Oluwatobi Aluko photo by Praise Samuel
Show Notes:
2:50 Tobi Aluko–Nigerian copyright
5:30 Nigeria’s national AI strategy and creative sector
7:25 Emily Gould – query on tech sector’s strength in Nigeria
9:40 debate over UK’s copyright provision for machine-generated work
11:40 Johann Brandstedt - Sweden’s approach to copyright
13:40 Czech Republic case on AI and copyrightability
16:40 Aluko – prompting versus creating art that involves effort/process
19:00 US copyright for A Single Piece of American Cheese
20:30 Prof. Ahmed Elgammal - background of AI use in the arts
23:45 19th C. French case on copyright of photographs that show originality
24:45 US refusal to copyright AI outputs
26:35 Bartz v. Anthropic and Kadrey v. Meta
30:45 AICAN project
32:55 GAN-based art movement celebrated prior to prompting
34:25 artists survey on value from AI
37:15 backlash of Christies’ auction of Generative AI works
41:50 Elgammal – effort versus intentionality e.g., DuChamp/ready-made art
42:30 intentionality and agency missing with prompting and random output
51:40 AI's cultural impact
1:04:30 Beethoven X project
Please share your comments and/or questions at stephanie@warfareofartandlaw.com
Music by Toulme.
To hear more episodes, please visit Warfare of Art and Law podcast's website.
To leave questions or comments about this or other episodes of the podcast and/or for information about joining the 2ND Saturday discussion on art, culture and justice, please message me at stephanie@warfareofartandlaw.com.
Thanks so much for listening!
© Stephanie Drawdy [2025]
I'm all for airing for good and not airing taking a bunch of stalling creativity.
SPEAKER_04:Welcome to Warfare of Art and Law, the podcast that focuses on how justice does or doesn't play out when art and law overlap. Hi everyone, it's Stephanie. And that was artist, lawyer, and data protection specialist Toby Aluko from a recent AI and IP panel discussion. What follows is a recording from that discussion in which Toby is joined by Johan Branstead, an IT and publishing specialist and artist rights advocate, and Ahmed Alkama, an AI researcher and creator, recorders professor, and director of records art and AI, and the founder of Playform AI, a platform for artists to integrate AI into their practice. These panelists are also joined by Institute of Art and Laws Assistant Director Emily Gould for a conversation in which we look at the issues surrounding AI and IPA from a global perspective, from Nigeria and the UK to Sweden and the US, considering both the concerns and the benefits of emerging technology in the arts. Welcome to Warfare, Heart and Law and Second Saturday. Thank you all so much for being here. It is such a pleasure to bring together this panel of individuals who have a deep understanding of the issues around AI and IP. I will give a brief introduction of the panel and then we will jump in. We have Toby Aluco, an artist, lawyer, and data protection specialist. Thank you, Toby, for being here.
SPEAKER_02:It's a pleasure to be here.
SPEAKER_04:Johan Brandstead, IT and publishing specialist and artist rights advocate. Thank you, Johan, for being here.
SPEAKER_03:Thank you. Pleasure to be here.
SPEAKER_04:Ahmed Algemal, an AI researcher and creator, uh, Rutgers professor and director of Rutgers Art and AI Lab, and the founder of Playform AI, a platform for artists. Thank you so much, Ahmed, for being here.
SPEAKER_00:Thank you for inviting me.
SPEAKER_04:Together with this panel, we will launch in. And Toby, perhaps you could begin. And earlier in 2025, there was a new AI initiative in Nigeria. So I don't know if you want to give us a bit of background on that and then tell us what your thoughts are on the issues and bonuses that you see coming out of that for artists.
SPEAKER_02:Right. So Nigerian law on AI. So in Nigeria, our copyright act hasn't been amended. The way we see copyright in Nigeria is that it has to be the work has to be created by a human being. That's the way to get copyright, or the work has to be created for a corporate body. That's like a company. That's how you get copyright protected. And also in Nigeria, to get copyright protected, once the work is created, it doesn't have to be registered. Once it's originally created, it's you own the copyright already. That's the way it is in Nigerian law. But the Nigerian Copyright Act has some um, will I say loopholes or the way it's drafted? It says that there's a part in the Copyright Act that's section two, section two of the Nigerian Copyright Act says that for a work to get copyrighted, there has to be some effort that has been expended on making the work to give it an original character. So I was looking at this from the AI perspective, because I know in America, um to get AI, AI is not, AI can't own copyright. But I think there are some cases whereby where there's a mixture of human effort and AI effort, you could argue if the human effort is more than the AI, if AI effort, you could argue that such work can be copyrighted. So in Nigeria, currently, it says that if there's some effort that has been shown into a work to give it original character, such work can be copyrighted. So this makes me think that if we use AI to create a work and you also put some human touches to it, would the work pass as copyrighted? Would it be copyrighted? And well, there have not been litigations on that yet to determine what the cost would say if you create something with the help of AI. But my view to that is that AI, AI's work at the moment, can pass as copyrighted if you can throw some effort as a human being, human touches into it to create it. So that kind of creates um, like I say, a loophole for AI work to be peripherated in Nigeria. And um at the moment, even though there's no litigation to see what the court says on it, I think that as it stands, AI work could pass if you put some human touches into it. That's the way I say it. Then also there was like a national AI strategy that was put forward by the government. And some of the things I read on it was that with every situation whereby some works or some jobs that that could be done by human beings would be taken over by AI in the future, is it going to affect like the economy from the perspective of job losses? So these are like um issues that are still new to the country and which we're still watching to see the way the whole landscape evolves. I think that's my perspective on it.
SPEAKER_04:Thank you so much. It's very interesting political issues that are swirling around how AI initiatives are going forward. For Nigeria, what I'd read was that they do have this desire to be a leader of the African countries in its AI initiative. And so I was curious, uh, because they're saying they're getting all these different perspectives from the different sectors. What do you what do you think?
SPEAKER_02:So in Nigeria, the creative sector is one of the leading sectors in the economy. Um, because Nigerian is known for, in Africa, for movies, um, music, and art. So the way the law stands, uh, because the act hasn't been amended to take into cognizance um what AI can do. So my view to that is that um I think um maybe litigations have to come forward so that the courts can say whether could interpret that aspect of the copyright law, which says that you just need to show some effort of originality for work to have copyright abilities to be copyrighted. So I feel that um my view on it is that we need to wait to see if um cases will come forward and sections of the copyright act could be interpreted, given reference to the current phase we are in with AI.
SPEAKER_01:Just sort of thinking politically and in terms of the different voices that um are you know are involved in this debate, whether the tech industry is strong in Nigeria and whether there are AI developers who are maybe lobbying for more sort of liberal rules which would allow them to have freer use of existing copyright works, because I know certainly in the UK that has been a very sort of powerful debate, and I don't think there's resolution yet between those two sides, which to some extent are are sort of pitted against each other. I mean it's obviously more complex than that, but um that has certainly been uh sort of the essence of the debate in the UK so far, sort of between the tech side, the developer side, and the creative industry side. So I was just wondering, Toby, with the saying that the you know the creative um sector voice is is really strong in Nigeria, whether there is you know an equivalence strength of voice from the tech side.
SPEAKER_02:Yeah, in the tech sector, we haven't really had like uh would I say companies like OpenAI or in Nigeria currently, we have like there are tech companies that use AI to create um to for their services. And but in Nigeria, um there's been a push for, would I say, from the part of like personal data, protection of personal data, there's been the push for stronger laws and the way you use personal data in businesses, the way you use personal data in in the works that creatives or would I say technological companies do. So we don't really have the same, last I checked, it's not like at the same level with the UK when it comes to tech companies pushing for um stronger, or would I say lenient um routes to create their technology and for lesser um lesser regulations from the from the government. So at the moment, we have like a national AI strategy in Nigeria, and I think that's where last I checked, that's where it is currently. There haven't been issues of regulators against tech companies from the side of AI.
SPEAKER_04:Just a wrinkle I would bring up too, since we're discussing the UK as well, the issues that the UK is seeing with extending copyright to machine-generated works and whether or not that might come up if indeed other countries like Nigeria start to do that. The idea of how you define authorship and originality and how much of a disconnect there might be if you're gonna describe uh extending copyright to an original work, but then is the original work created solely by a human in your in your interpretation of your copyright act versus extending that to machine-generated works? I don't know if anyone wants to jump in on just that point.
SPEAKER_01:Yeah, yeah, I think I think just from the UK perspective, um, I I don't know if that was what you were alluding to, Stephanie, but there is um quite an unusual provision whereby there is a specific copyright for computer-generated works in the UK Act currently, although that is very much up for debate. So the consultation that the most recent consultation that the UK government published um suggested that that you know that debate be opened again. And is that a right that is used very much? And is it appropriate? Because obviously, you know, our our copyright act dates back to 1988, well before AI was anything like at the state of development it is now, and so I think um it's possible that that right might go altogether, um, which would put the UK in a much more comparative position with most other countries. I think it's quite unusual to have that provision. Uh, I mean, lots of countries which derive their laws very closely from UK law don't have that necessarily. So um, yeah, it's very much an open debate, I think.
SPEAKER_03:Yeah, I think in uh in Sweden we have a similar uh principles to Nigeria. It's you you own it what you make by default, um no registration needed, and and that's kind of where that's the first major clash, right? If you need permission first, then we're in a totally different space. And then when it comes to authorship, I believe it's uh UK, India, and Ukraine, and possibly a few others that have these special cases where you can actually delegate most of the work to the computer and have it made for you and still claim copyright for it, which is very odd. Uh it's odd with the basic principle of what creates value, right? It's the human authorship that's the taxable income and the the value add. Uh, whereas you don't want to place incentives around automation, usually, right? You that's not what we want to create incentives for because it doesn't build a taxable uh income base to to basic data on, more or less. So I I I think it's super interesting to to compare these because uh people usually don't reflect a lot, and then Sweden is in a lot of cultural ways and a lot of other ways a colony of the US. Popular culture is inundated and also the tech companies, there's a very close links between the US and and Swedish Sweden. But but some of these issues really brings to light the differences at the very foundational level. We're talking about the constitutions of the states being built from the ground up very differently. Um but I think one thing to we might throw in here because we're in the same situation as in Nigeria, that it hasn't really been tested locally whether it's what's copyrightable. But we do have a thousand registered hybrid works in the US and some principles emerging from there, which sort of align with, I believe it was in uh Czech Republic we had the first European uh court and the decision on copyrightability, and they pretty much went with the principle of well, you didn't make it, so you can't claim it. What are we even discussing here, right? Um, but I think one test that's worth throwing in here is one that I came across a lot. So I have a background in in uh comics and licensing, international licensing, uh, from Japan to Sweden, from Sweden to Germany, from US to Sweden. Um and uh those are very often joint authorship works. You have a writer, you have an artist, sometimes you have a bigger team. Um, and there's the principle of starting medium, which I think is a very good one to bring in uh to get the distinction between prompting versus other ways of hybrid process to make an image, right? And the the principle is very simple: you can have joint authorship and shared copyrights in a children's book that has pictures and words, but there's also separate, so the writer can't claim the pictures and the illustrator can't claim the words. So that's that's quite that's the test, right? And that's basically corresponds to prompting, right? If you started writing a text, you can't claim an image, uh, and that's really being kind of okay. So, how much control over an image can we have via text, even to begin with? Uh, and to me it's very clear because I've I'm a cartoonist originally and I've written comic scripts and I've also illustrated comic scripts, which means to me, well, it's obvious that you write something and then you get an interpretation, and this is entirely at odds with how it's being presented by the major AI companies. They want you to feel that you made what you prompted, right? And that's how everything is worded: the marketing, the services. It all says you created this, and even if you ask ChatGPT about copyrightability, it will tell 700 million people that yes, what you just prompted without doing anything else, it's yours to claim. Uh, so we have an issue there with common sense from the perspective of anyone working within the creative industry and the sort of mental models that the tech industry wants to impose, which is well, this is just a tool, it learned like people, and you made what you asked for. You made what you bought. That's basically the picture, and it's a very incoherent picture to anyone who works inside the field, but to all amateurs, it makes perfect sense. So just thought I'd have your opinions on this because this is something that uh really bugs me and I fight about every day. Would love to hear hear your your thoughts.
SPEAKER_02:Yeah, for me, I feel um when it comes to creating like art, because I create art myself, I do paintings and I think prompting and creating art is totally different. They're totally different things. Like prompting, you don't it doesn't take the effort it takes is just to imagine something and it comes into fruition. You imagine something and you type it, and it is different from learning how to create um art, which is usually the I think the beauty in creating art is the process of creating it, what you go through, thinking about it, putting it on canvas or on your iPod, because I do digital art on your iPod. It still takes efforts to create. So prompting and creating to me are two totally different things. And I feel I I I tilt towards the um position that those shouldn't be copyrighted, that prompting should be shouldn't be copyrighted because it's it's going to lead to dilution. That's how I say dilution of works. There'll be like proliferation of what you can't, you can't really tell whether it's created by a human or it's created by machine. So the law dragging its feet on whether machines should have copyright. I think it's fair. I feel it should follow that way, should continue that way till we totally figure out what we're gonna do in the end, whether machines are gonna have copyright in the end or it will forever be humans. And ITO was the human expert. But the loophole, like looking at Nigerian law, which says that some efforts, since that hasn't been tested, doesn't mean that if you start with um uh with you prompt an image and you paint over it, would you have copyright for that? Like you prompt something and you paint halfway through. That's that's something I think of. Should you should you earn copyright for machine-assisted work or should it all just be human?
SPEAKER_03:Yeah, the hybrid processes are like the where it's really at right now, where we had this uh, I think it was called American Polone or something like that. Uh was a work of art, if you want, uh legally a work of art. Uh it was basically prompted fill in the blanks. You uh they they sketched out the simple drawing and then had uh generative AI fill in the blanks, and they got uh what some US legal experts call a thin copyright or a slim copright, I forget which. That says, well, the human-added parts uh are for short cooperative, and there's really no way to set of separating them. So it's kind of a proof of concept of minimum viable effort to claim uh an output uh by way of and and here's the interesting thing is I think they you're actually in the starting medium. So the principle of starting medium, if we kind of okay, so it starts with an image, a blank canvas, but it's really minimum viable effort. Um yes, it kind of nominally you've composed the work, you've decided where things go in an image, but from there it's five minutes of work in total for something that looks like it could have taken a month, and is probably based on something that took months or weeks or at least days. Thoughts on that one? Have you seen it? Uh in another one?
SPEAKER_02:I haven't seen the baloney.
SPEAKER_03:I haven't seen the I forget if that's the exact title.
SPEAKER_04:The American piece of Swiss cheese or American a single piece of American cheese?
SPEAKER_03:Yes, yes, that's it. Bologna, that's my that's it's all baloney. Oh I have a single cheese. A single piece of American cheese. Yes. I missed that. Yeah.
SPEAKER_04:I would ask Ahmed, what are your thoughts on all of this that we've been discussing from your perspective?
SPEAKER_00:Yeah, um uh first of all, I mean, um I would like to remind everybody um and the listener that uh the use of AI in making art is not new. It's have been around for as old as AI itself, more than 50 years uh now, where artists have been some artists have been pioneering exploring AI in their own work. AI itself has evolved a lot, and what we see right now in terms of prompting uh is just the last, the latest wave of AI, which creates a lot of problems. Uh but just if you go like five years uh ago, before that wave started in 2022, uh many arts have been using AI uh um around 2020. Um uh we introduced Playform, for example, as a platform uh at that time, and many arts have been using uh AI at that time, before prompting, and it was a very different process because you um can train your own AI from scratch based on your own images that capture your own work and style and create art, a very uh creative process. Uh and then uh came prompting, and and all these issues came out with prompting. Um, fundamentally, because um, when you prompt um and you create an image in a second, and you think obviously that uh you created this, uh, that's you have the impression that you create this because you write this amazing prompt. But I mean, suppose somebody else comes with exactly the same prompt. Um, so you have two people come with the same prompt, and and uh you're gonna have uh probably uh most probably gonna have very different images. But why uh the reason why is basically because um in the back uh end of the AI, there's a random number generator that just gives a coin and uh using the same prompt you end up with different things. So the the the agency here is really um uh missing in terms of how the word translates to an a final image is a black box in the middle. That that black box has been trained on trillions of images and involves a lot of randomness engineering the image. So the agency between the input and output is totally uh missed out. Um so that's actually the reason why um the copyright uh office, uh particularly in the OS, have rejected several cases of uh uh giving copyrights for uh AI generated uh artworks. And um, however, this uh I don't think this will stand for forever. I I it all uh I believe that if somebody cab was a case where um they could use AI to create uh not necessarily prompt, but but the process really involved their own work and their own control, the copyright officers might accept that. I haven't seen any cases like that yet. But the thing is also that uh that um uh there are many things in my head here, I want to get straight. Um uh that there's a in the 19th century, uh the history of photography, there was a very similar scenario where um courts in the beginning uh rejected to give copyright to photography. And there was a famous case uh around 1861, 62 in France, where um famous photographer Mayer and uh Bierson wanted to get copyright for their work and the court rejected in the beginning. And then after a lot of debates, they finally gave copyright to photography. And the main the fundamental reason was uh um that the artist has to show that their own um their own uh identity is reflected in the photography that they are making. Uh what's different between me taking the same camera and and having the same shots? Why gonna be different? If I if you can show me that there is something about my artist's identity that reflects in the art, then copyright can can be granted. And that's what's missing now in AI. Uh, how to show that your art's identity is there when you are using AI. And I guess that depends on what kind of AI you are using. And um the uh when you look at now uh the the courts and and a copyright office, their standing in this issue is totally confusing. Because, in one hand, um uh there have been lawsuits against um companies making AI um uh art platforms uh claiming that the their work is infringement on copyrights, and the courts have been rejected in all these cases so far, um, uh basically saying that copyright uh generation generation is transformative enough and is not uh uh copyright infringement, which would imply that the generation is somehow original. But at the same time, the copyright office have rejected any attempt to copyright this supposedly original work uh because you are not infringing. So that's kind of uh kind of uh contradicting uh situation at this point. And and most recently, um from June this year, uh there have been cases where um uh um uh the court said that basically uh training AI on copyrighted material is a fair use. Um and that basically makes uh all this company training AI using other people's uh material um not liable for copyright infringement, it's fair use. Um that's also a very uh odd situation for artists. I think now that situation is very uh uh muddy and and against artists in all cases. You cannot use AI to get anything copyrighted. If anybody else uses your AI um to generate uh to easy AI to generate things that look like your work, you are not protected. And if companies use your AI to train, uh that's fair use. So it's a very bad situation for arts at this point.
SPEAKER_03:Yeah, and if we look at the Bards versus anthropic, which is the one you referenced, I think is super interesting that two days later in the next district court over, uh it basically got scolded by another judge saying, Well, your reasoning is entirely flawed. You uh and the entire premise of the case was what I like to call anthroxic, uh meaning toxically anthropomorphic. Uh so it's a portmanteau of like you bring in this human, you try to humanize the system and make people think of it as a human in order to establish a mental model where the machine use is comparable to the human use. So there I believe what Bart's judge in that case is what he said was well, it's just like teaching a class of children to write, and that's it, that's a good, right? Um, and that that's not an issue competition-wise. Uh and that's entire reasoning was demolished by the next judge to over. But since the plaintiff came with the anthropic and served the anthroxic framing from the get-go and said it learns like people, then all the judge can say is well, it learns like people. They didn't do that in the next case over, but there they failed to uh demonstrate market harm. And I watched the session, it was really painful to see because the judge was basically begging them to bring some evidence of having suffered concrete harm from the plaintiffs in the room. And I say, I understand the issue, it's about the next Taylor Swift, it's about budding talents. How do we make incentives for them? It's about market obliteration. Um, that I believe that's the term he used. Um, and I would love to give you right, but you're not presenting the arguments, and I can only go by your arguments. So some commenters said, well, he basically served the blueprint for how to argue against fair use in the next ruling, but that won't happen this year. Uh it might happen next year. Um, and he still ended up saying, well, it's still transformative fair use based on these arguments, right? Based on the facts in this case. Um so I think we're we're kind of inching closer to the core issue of this mad frame of comparing products to people and making assumptions on morality and ethicality and legality based on products are like people. Uh but it it might be years yet, and we're the markets is being obliterated in real time. So artists are basically being bled out as this plays out in court. That's my view. It's it's bleak. Sorry.
SPEAKER_04:Thank you. That second case, uh Kadre, I would just point to, I believe Toby had raised earlier this issue about market dilution, which all came up with Judge Shabria's opinion in the Cadre case. And as Johan put out there, yes, it is like a blueprint going forward for plaintiffs to hopefully successfully argue against transformative fair use. But yeah, this is definitely a concerning space that we're in, at least in the states, in regards to the way the cases are unfolding. Going back to you, Ahmed, I wonder for all of the different projects that you've been doing. Doing, knowing that this is this kind of situation where you can are not able to uh receive copyright protection. I wonder what your thoughts are about whether or not there is an issue of creative motivation or lack thereof.
SPEAKER_00:Yeah, um frankly, I mean, when I I did most of this project, um um I didn't have that issue in my mind at all. I didn't, I mean, as a as a research scientist or as uh an artist also um approaching this project as a creator, uh I really didn't think about that. I I'm not doing that to get copyright, and I'm uh I'm not getting copyright that does not prevent me from doing them. Uh that's fundamental. Here's uh also a very interesting uh project that relates to this. Uh back in um 2017, um, we have developed an AI algorithm that uh looked at um a history of Western art and tried to generate images that uh learn the rules of art, but at the same time uh don't uh follow uh any any art movements, try to generate its own style, basically. And that was called uh AI can ICANN. Um and and uh it made a lot of uh media attention at the time, back in 2017, that was way before uh the current generation of AI art making. And the headline at that time was basically finally, things like finally AI can generate something that we like, the future of art is in danger, things like that. Really a lot of debate about uh uh AI stepping into the creative domain and can create something uh creative and novel um um and and not never seen before. Um uh one thing interesting here is that um we applied for a patent for the algorithm, and the patent was accepted. So we officially have uh um uh an algorithm that's patentable, the S patented that create art by itself. In the same time, copyright office that's often uh issue the same patents have been rejected any copyright um for AI generation from artists. So that that uh for me it's a bizarre situation because uh you kind of patented an algorithm to generate art, but in the same time, you we didn't never try to claim copyright for the generation, but it would be funny if we uh asked uh copyright uh office to um for a copyright for the generation and how how that would be seen given that the algorithm itself is patentable. Um so it's it's uh the all these issues came later. I said as I said before, uh I mean there was not much debate about that until really the the Brompt-based AI came around in late 2022. Um before that, uh I mean there have been five years of uh at least of AI generated art uh movement called the GAN-based uh art movement, where there was actually AI art was have been celebrated. Uh there have been AI art exhibitions in museums, in in art fairs, in galleries, and uh it was very, very healthy, and and people really liked that kind of art, um, AI art at that time. And I said that there are a history of that go back 50 years uh older than this. All this debate comes really with the prompt-based AI for the two reasons I mentioned earlier, the fact that uh the prompt-based AI really comes with trillions of images of uh uh that are copyrighted, baked in, and using to train the system, and that changed everything. Once this came around, we start seeing basically backlash against AI, and we start seeing AI is being banned from artist community and and uh uh lawsuits started happening. Um uh, and really that I have an article at the time uh in I remember in Artnet News or something uh that really said that basically the AI art era is over back in 2022, because really the the main amazing things that happened in the five years prior to 2022 was really an art movement uh that where artists use AI in amazing ways. Once 2020 came around and prompt-based came around, it's no longer art. It's just basically making images using prompts. Uh uh to make it art, to to qualify as art is a totally different thing. Um so um, and also I have been working with many artists in that era um until now who have been using AI in amazing ways. And and we we have a paper where we actually um we surveyed many artists who are using AI in around 2020, again, very early, and ask them um uh uh why use AI, what do you find as a value when using AI? And basically there are two things uh artists keep repeating uh in terms of what they they say. First, um the fact that uh AI give them um um um creative um uh ideas, the fact that basically um you can plug in your images or ideas and and can give you something that you never thought uh before about. Because AI basically doesn't have the same constraints as us when looking at images. And that again was before the prompt-based uh scenario. Uh, because uh once we add prompt to the problem, now you kind of render the image with the lens of the language. You are constrained only to what can be described by language. Um uh and that's a limitation actually. Before the prompt-based, uh you don't have that limitation, you can re-render the based the pixels based on your own images without any um uh uh limit of the language. Imagine you are working on abstract uh uh uh art and and and you cannot describe what you're doing in with any words, and and AI now is limiting to you to do that. Uh that's so that the first thing is that this that the fact that artists find AI really give them creative inspiration. And the second value was um uh that uh artists always say that AI gave them the creative volume. Um the fact that AI can help them in creating lots of and lots of uh assets and material that they can use in their own project or or they can use into whatever they are making, whether they are making video or making a mosaic or making uh whatever project, the fact that um AI can work as an uh uh studio assistant for you uh that you can uh you can use for that verbos. Um so again, I mean, when artists use AI at that time, they didn't think much about uh the copyright issues because mostly they were using AI based on their own images, not on somebody's else's images. And secondly, they are using AI as part of their own process. It was not really uh something alien to them, it's part of the process. I'm using AI as any other tool that I'm using. And again, that whole things changed where you are just using prompts and your whole process is just input a text and getting an image, and and and and uh now it's really uh problematic in in many ways.
SPEAKER_01:I think um the there's a really good sort of example or illustration of how things change very much with prompts, but so generative AI tools. Um so there was a there was a Christie's sale of AI works. Um, I can't remember exactly when it was a fair while ago, but after sort of 2022. Right, okay. So you know the one I'm referring to, and there was a real sort of backlash within the artistic community, and artists got together and asked for the sale to be cancelled, but then there was kind of a backlash to the backlash by a lot of the artists exhibiting because they said exactly as you were saying, Ahmed, we are using AI as this creative tool, and we see it as a way of you know progressing our work, moving moving uh the artistic canon forwards, if you like, uh within the digital art sphere. And so I think that I think that does illustrate really sort of clearly the the sort of different um approaches and the way you know there isn't kind of a one-size fits-all when we're thinking, I think, about copyright. In we can't sort of generalize, I suppose, I don't think, really, when we talking about talk about AI generated art because it can be so very different, you know, it AI might be just a simple tool for somebody using a um a generative AI algorithm versus you know artists who are working as artists all day, every day, and using a very different kind of tool that they have created by using often their own work and you know by sort of curating that tool in the first place.
SPEAKER_03:Absolutely. Yes, and I think the backlash was about kind of this piracy-based uh image models being fronted and invading the space very obviously from the get-go. And I I really think they did themselves a disservice by hosting this and and muddying the waters by mixing up people like Rafi Canadol, who has done this custom software that makes custom living images, and it's very much bespoke and the uh artistic process, versus someone who prompted something in stable diffusion and then stitched some pictures together. Um, and I guess I mean the you can still copyright the work that's based based on stolen paint or stolen images, um, as long as you add value to it. But I think the red flag for red sheet for a lot of people was those specific image models being used. Um so yeah, there was an interesting discussion after. And I I I I'm totally on on board with protesting against it because it we need to debate and get to the point to the core of what are the real issues here. And I actually debated a bit with one of the guys who had uh an artist who makes pseudophotography, calls it post-photography, uh, where he uses these kind of models to make something that's very clearly photorealistic, but also very clearly not photos, because things break apart at a closer look, and he kind of uses that consciously. Uh, whereas the issue is taking the shortcut and never mind the details, and just getting the work done in the smallest time possible. So he's putting himself in a pretty tricky spot by using these things, but I think the position is defensible because he really uses the tools and the content and the process in a very conscious way to specific ends uh creatively. So yeah.
SPEAKER_02:Yeah, for me it's uh I'm also on the view of about the process. Like art is not just about the final image, it's about the process and how you make it. So, like um Ahmed said, prompting prompting is actually a big issue because I don't agree that uh putting images, but what if you do a bunch of prompting and you gather your prompts as like a mosaic? Sort of is could that pass out some effort to create art if you um clip different images you've prompted together and as a finished work to create something? That could pass out some effort, or I don't know what do you think about that?
SPEAKER_00:I don't think I mean the effort should be the the metric here. I mean, uh we are a hundred years now after uh Marcel Deschamps have his his fountain uh as an art, basically the unary thing that he did as a sculpture, and and that opened the the whole era of ready-made art, basically, when you can use an object and and declare an art. And he had a point basically that artist can can make anything and call it art. That's how his whole point. And so the to come now 100 years later and and still talk about the effort and the time spent as a as a metric to qualify something to be an art or not, I don't think that's right. I I think uh the main issue here is really um um uh intentionality, obviously, is one important thing. Um, and definitely Marcel Scham had the intentionality of making art this way, uh to make a message. Uh, when you're prompting, uh the intentionality is broken. Uh part of it, yes, you you you write a prompt with intention of something, but you actually get on the other end is something totally up to the uh to randomness in the process. And and uh you might like it or not, you might start tweaking the prompt to try to control it to what you want. But most of the time you end up with something that you like. So your role is more like a curator. Uh, you start with with some idea, you got something, it's not really what I like, but I like what I get. Uh, and and here it is. I mean, uh, so that's what's missing in the process, the fact that uh the agency is is not complete. Um uh part of it is random, part of it is is uh is mashing up millions of images uh based on language to generate something. Um but it's not about the effort because it can actually take you a lot of effort to create some image with a prompt. It can a bit of trial and error uh to to try to really control it because we are very hard to control. So the effort is there, the the agency is not there, the intentionality is what's missing, or or or not 100%.
SPEAKER_03:Yeah. One of the examples I like to go back to is uh I mean, generative art has a prehistory to digital art, right? There were spiral graphs and there were paint spinners and people experimenting with that in the in the 60s to figure out some of these things, right? Um, and the principle of starting medium is a is a good one, but one that came out of that whole line of experimentation where they did things like okay, so here's a paint spinner, we invite the visitors to pour in the paints and then uh press the button or spin the wheel, and we get an original picture and we put it on the wall. And then you get the same core question that you have today: who made the picture? Is it the guy who made the machine and patterned the machine? Is it the user or is it the machine? Right? Who is the real author here? Uh and the principle of independent origination came out of that, if I'm correctly informed. And that says, well, if a layman looks at the pictures on the wall and he says, Well, I can't tell apart that these are being done by different people, then there's no basis for a claim of human authorship because there's no originality in it. Uh, and and it's kind of two layers to it. One is can you tell it apart at all? And then the next layer is can you tell apart that specific people have made this? Because if you can't, it doesn't express the personality of the author, right? Um, so it's one is yellow and one is green. That's not enough personality, even if one person really likes green, right? Um, so I think that's a good one. And also if you look at those three, like user, machine, and uh and the machine maker, uh there's a really good paper, I forget the title then, but that really went to the bottom of from a societal perspective, if we look at copyright as an incentive machine, an incentive framework, if we place the incentives around the user's effort, the machine effort, or the the machine maker, like let's just look at what happens if we give output rights to the guy who made the machine, to the machine itself, or the user. And the conclusion out of all this was well, it's not about who, it's about we don't want to incentivize automation, we don't want to incentivize volume of empty work. We want to actually incentivize the process, as you mentioned, Toby. That's where the value is created. Um, and also as a matter of scale, which they it's a few years back and they kind of predicted where we are now, is that if you incentivize empty automation, then you get system-wrecking volumes. So it doesn't really matter who gets to claim all of all of these outputs, if there's an incentive to create more and more and more, you get too much. And market dilution is one thing, but there's also a democracy aspect and giving up on truth altogether and telling what's real. I mean, it's a it's it's a much larger issue than just the incentives around art. It becomes can we even navigate reality if we're surrounded by fake pictures? So yeah. Um, and I to your point there um about the patterns for the machine, that to me sounds a lot like oh, I forget now, but uh the gates of paradise is the name of a work that's entirely machine generated. Um, so there was no human in the loop at all. It's a randomizer that produces uh generative works. And the guy uh I think it's Stephen Thaler, right? Is on the mission around the world to get authorship rights for machines. So that's a I see either he's a lunatic or he's kind of pressure testing the legal systems around the world. Either way, that's how it ends up working. It's it's how it does the system hold. Um and I believe they assigned authorship rights, I believe maybe in South Africa. Um, and I forget now because he did a patent machine and he did an art machine, uh, and he's trying to register the outputs of both of them around the world, right? Um, and I believe he actually managed to get one of his inventions or artworks patented or copyrighted, I forget which one. Uh, but they ended up not designing the copyrights to the machine that he asked for, but to him, right? So the author of the machine got uh patents to the outputs, which is an anomaly, as I understand it. Um but yeah, uh so but I I guess one of the uh courts when they assessed uh his attempts to register the gates to paradise, the randomized works without a human in the loop. It was uh well, we already incentivized you, we already provide protection. You're the patent owner of the machine that you made. That's your incentive, right? You don't need additional incentives in claiming also the outputs of the machine for the reasons discussed just now. So I think that's how how that story ended up for them. But as I understand it, this guy goes around and tries to escalate. And you have Jason Allen also doing his space theater opera. Uh, one of my favorite kind of one-liners from one of the court rulings when he tried to copyright this mid-journey image, which looks like a generic mid-journey version 3 image. Um, is like, well, you're not an artist because you found a pretty stick in the woods. It doesn't matter that you searched for 600 days, you still didn't make the stick, right? Um, and that's it's a bit mean, but uh there's a core of truth to it, right? And and I believe that one was okay. One part was he did all these iterations until he found something he liked. And the next part was he did some minor edits, and I believe they assessed these edits to not add any value, uh basically, it's marginal adjustments, and they say, Well, that's not enough. You didn't add any creativity in that step, you just tweaked it a little bit in the edges, so no copyright for you. That's I believe so. Those are some of my favorite cases to track Steven Tyler and Jason Allen. And um, and they teach us a lot about like what's the underlying principles here uh that we're finding out as we go on.
SPEAKER_04:Also, that whole layer of the training questions that go into mainstream models, like the mid-journey models. What are the infringing aspects of those outputs? Like, that's a whole other conversation that's really not even being factored in many times when they're having that can we copyright an output question. That's what I love about models like Ahmed, like what you have been talking about, where you can train your own model. And that to me, it really does speak to an ethical approach to this. Well, unless there are any questions about that, I also was just gonna dovetail into this question that Ahmed had raised as a point about cultural stagnation, which I think everything that we've been talking about kind of smacks of questions about where is our culture going. So, anybody want to jump in on that?
SPEAKER_00:Yeah, let me start. Um, um uh yeah, um I think uh since um 2022, um Chat GPT and and uh all prompt-based image generation came around, um, definitely the line between what's real and what's fake become very blurry. Uh so since then we have seen lots and lots of um images in social media that are fake. Um every day now it's uh being generated. Very good quality, obviously. It's very hard to tell videos now, real videos from fake videos. Uh right after Shad GBT came around, there was a nice story, funny story that um Amazon um uh there was many Amazon books, uh bookstores uh online just making fake books and selling well, actually, these fake books. Um uh so uh really the the the there are two big big problems with that, obviously. The problem of um that the line between what's real and what's fake is totally blurred at this point and gonna be more blurred to uh to in the future to the point that we will not really know what's real and what's fake anymore uh in any medium. And the other thing is that all this fake stuff is gonna be uh um by scale, gonna be the majority of content on the internet and and and and and everywhere. And that means that the next wave of AI models will be generated with more of AI generated images and videos and text than human-generated ones. Uh, because one of the problems with AI now is that the lack of data already digested all the internet, so they were looking for more synthetic data, more AI generated uh data, you're welcoming that to train and skill their model. But then we have a problem that now we are uh going into culture with technicians because uh the majority of content there will be AI generated. Um, and um what that do to our culture, that's really um a big problem.
SPEAKER_03:Yeah, the if I may, there have been this discussed to one of my pet peeves. That that's again like what when we see the mainstream media reporting on Sora, for instance. Uh there are a couple of framings. Uh one is can machines be held liable? Like, is the bot to blame, is the AI to blame? That's one of the framings, right? People are really struggling with this, and the other is users are using this tool to do X, Y, and Z. And to me, that's the other crazy frame. If you have the maker and the machine and the user, right? One is about blaming the machine, one is about blaming the user. But the user can't make this, and the machine can't make this unless there's a service provider providing this live service, actually producing the work, and nobody points in that direction. Uh, and that's it's baffling and horrifying, frankly, to me, that nobody says, why aren't these media production companies being held accountable for what they produce? That's absent in a lot of the discussion. And I I just don't understand it. And the the best conspiratorial thinking is they have been really good at PR, at establishing some frames of thinking around uh liable human machines and about users getting tools that make that they use to make things, uh, and avoiding the obvious thing that is tech companies moving into media production. That's what they're doing, right? Except without the input costs and the output liabilities. So they manage to kind of have this old mental frame of these toolmakers or these neutral distributors, despite now being producers. And so this discussion becomes one of the DMCA, like, okay, so the platforms can't be held accountable, right, for what flows through the platforms, and we've already seen what happens if it sucks all the money out of media, the traditional media, and everybody else out of their personal data, and you have an ad monopoly and a social distribution, media distribution monopoly, right? Um, and now they're moving into to even compete directly against legacy media by producing their own media based on theirs without paying for it. Uh, and we're still stuck in thinking about them as tools and as distribution platforms where really they are production houses now, uh, fully automated production houses. Um, yeah, uh, so I I think just placing those the product liability and the service provider responsibility in the right place is where we need to start if we're going to get anywhere with this.
SPEAKER_02:Yeah, I think my view on the my view on the culture of stagnation is that it can also take incentive away from because when there is like proliferation of AI creator-generated work all over the internet, if you're like a graphic designer, it's it takes jobs away from graphic designers because now you could just use uh an AI app to create graphics and you no longer need to pay graphic designers. And it kind of takes away incentives from uh skills like skills like graphic design, like because if you could uh just use uh AI to create your to brush up your images, you don't need to pay photographers to take image photos of you. You can use AI to uh brush up photos to make it clearer and fine. Now you could use uh these apps to create designs and uh it kind of kills that's what I say. The incentive for people to learn some of these skills that we've been uh doing for years, that's how I say.
SPEAKER_03:Yeah, absolutely. Uh and to me again it it's uh it goes back to the framing. Uh, if we think that an AI user is doing the same work as the one who produces the original work that's being used as a training material, or if we think of the machine as the actor, then we we miss the aspect of well, no, these are the service providers, the now media production companies competing directly against the artists and graphic designers and writers and translators and all of those, right? Uh but they're dumping the market, they're not selling these services at nearly the same price level. So when OpenAI produces a cartoon image, uh it does it in seconds, right? Fractions of a second, and it competes directly against the work of whoever made the original drawing who needs hours to finish the same thing, and they sell it for cents rather than selling it for the well, maybe a hundred bucks or whatever they would need to charge to to make a living off of making cartoons, right? Um and one part of it is automation. That's fine. I mean, it it does the work faster, but the other is sourcing, right? Uh the cartoonist needs reference material, they need production assets, they need tools, they need all of these business inputs to produce the image. And this trillion dollar company has the competitive advantage of not paying for any of it. They get tax breaks for the data centers and they they scrape the internet to get the production assets for free. So it's really an even uneven competition situation. And to me, it's actually the big picture issue is more antitrust and comp fair competition law than it's about copyright. Because again, they don't own their output, but that doesn't really matter. You don't need a copyright to sell anything, you can still. Sell bespoke images made in an instant, right? In a lot of business context, it doesn't really matter if somebody else has the same work, right? You you would go to a stock agency or you would license pre-made work anyway. You don't need the exclusivity. And you're not really using the picture as your core competitive thing. It's maybe maybe used for marketing or as a web page filler or whatever. It doesn't really matter if someone prompts the same headset girl for your support page, right? So yeah. So to me, it's it's uh definitely a case of Microsoft now competing directly against the Dartists they took from Microsoft and OpenAI, not their users, Microsoft directly.
SPEAKER_00:Except that um that um Microsoft and OpenAI does not at this point uh uh make entertainment and production their their main business model or or thinking this way AI and AI focus on AGI and Microsoft as a software company. Uh however, we have to remember uh how Netflix started. Uh when Netflix started, we were basically mailing CDs in the mail and we're mainly recommendation system for recommending you the next movie to watch. And we evolved into the major production company now of shows and killing all other studios. Um, so that's really um um how things went um uh for Netflix. Uh can we imagine major AI companies uh going into uh into the direction of media production and abandoning the idea of AGI? I don't see it yet. Probably there are some companies who will come up uh uh and focus on the production in particular. It might be again Netflix or or the Netflixes of the world where basically they are well positioned to use AI uh in production in the next level. But um the problem was that major AI companies have other missions they're trying to focus on, thinking that these are bigger business models to achieve, including automation of the world, taking the jobs of other people and tools and robotics and things like that. Uh so um they are not really focusing on uh that entertainment production angle.
SPEAKER_03:This is so interesting to me because I agree that that's how they position themselves and how they push things forward. But when you look today at what they actually do, if I have 20 bucks a month to spend on pictures, I can spend it on shutter stock, I can hire a fairly rookie illustrator, I can go to OpenAI. So even though they're aiming to be a tech and presenting as a tech company in practice, they are competing already today. Uh, and we see them also moving into media. OpenAI is funding an animated feature field now. Google is starting their own production house. So what we're seeing is that they're kind of shedding the pretense of being a tech company, they're spinning off this purely media-centered place, right? Where automation and everyone else's property as free business inputs is the starting point. And if you have that as your head start, then you can get pretty far and make some pretty bad movies until you get the hang of the rest. If you look at Altman and what he's doing, he says, Well, I'm making a religion, I'm making a cult, and I'm incorporating it as a company. It's all about getting people to think more about the future than they hear now. Uh, what might this be next year, right? What might might it be in two years from now? Rather than looking at, well, what are you doing right now? You've already wrecked basically academic research, um news media, and you know uh you're dumping the market for illustration and translation services, but it's always about looking into the future and seeing what's next and forgetting about this petty squabbles about the here and now. So it's fascinating to me.
SPEAKER_04:Thank you. One upside I see of our culture going forward using AI, I would just uh bring back to a Beethoven project that I learned about through Ahmed.
SPEAKER_00:I haven't heard about this. It's it was quite famous in Europe, not in the US, unfortunately. Uh so uh Beethoven uh uh made nine symphonies and uh he left some sketches for a 10th symphony, very rough sketches for a 10th symphony. And back in 2020, 2019, actually, uh Deutsche Telekom um commissioned me and a group of uh musicologists and a music historian to try to uh come up with a version of Beethoven 10th symphony based on the sketches and using AI. I was the AI lead in that project. Uh that was a celebration of Beethoven 250th anniversary. Uh COVID came obviously and delayed everything, but back in 2021, we made the concert um uh uh in Bonn uh played by uh Beethoven uh Symphony Orchestra, where we finished two movements uh of uh of uh the symphony using uh AI. It was really uh binding project uh using the same kind of technology that came late to be known as ChatGBT, basically the same language models, uh, to learn about uh Beethoven style of composition and orchestration, and now feed these sketches and try to see what kind of composition the AI can come up with. And and it was really a good project to show the potential of AI. AI was a building block in that project, but didn't cancel the human. I mean, there was a human musicologist, composers, they are historian, all the team worked together with the AI to come up with something, and it was a creative process where you use AI to come up with some continuation, for example, and then the composer would look at this and and approve it or not, or come up with other ideas, give it to the AI again. It was really a very good project to show how AI can be used as a tool in the artistic toolkit of uh an artist or an artist team to do something rather than just uh uh having a black box that you prompt and and you get a final answer uh that that uh really is problematic. So um that the the completion of Beethoven X is available on YouTube. You can uh find it and listen to it and enjoy it. Um uh it's it's it I mean I find it inspiring and in terms of uh um listening to a beast that Beethoven never wrote, but still based on his own notes. It's really interesting. You can tell it for yourself, search for it. Thank you.
SPEAKER_03:That's super cool, and I mean to not be the David Honor here being the uh public diffusion is an interesting project, right? Which starts from the same premise. We we'll starting with from cleared public domain images and making an image model out of that. That's still problematic for artists working in traditional styles today. If they manage to produce as good input, we still have an issue with automation, but we don't have an IP issue, right? Um, and it's uh I see a lot of potentially similar use cases where you can expand on our shared history. But again, you need to sort of have output watermarking or something like that, so you or or we will just confuse ourselves by inventing a lot of fake history, uh fake art history. But I think it's a it's a worthwhile project and it's interesting, and there are many likes it. Uh there was another one in Japan where it took the the grand old master of of manga, Osama Tezuka, um, and worked with his estate to digitize his entire archive and see if they could have it come up with new comic episodes in his style. But similar to what you mentioned, the estate had approved beforehand, and his old assistants uh and his old publishers were all involved creatively in the project to make sure that they they actually completed it with a human touch and in a good way. I don't think it was commercially successful, I haven't seen it, so I don't know if it was creatively successful, but it shows a good example of how to do it the right way. Everyone approving and everyone being contributes in a creative human way to keep a tradition alive. It's about building on what came before, rather than cannibalizing not only everything that came before, but everyone today as well.
SPEAKER_04:So yeah, and being transparent about the source that there was that AI element to it, then no one is misled.
SPEAKER_03:Absolutely. Yeah, the three C synatry, credit consent, compensation, that transparency.
SPEAKER_04:Yeah, yeah. Toby, what would you think? I I'm just curious because I I am curious to uh use an AI model built on my work, and and so would you or have you done anything like that? What are your thoughts on that?
SPEAKER_02:So I think that's good. I've because I also write fiction sometimes. I play with stories, and so sometimes sometimes I feed my um work into like a temporary chat GPT, the temporary part, and hear suggestions on tweaking or views. Like if I can get a human editor, I could use AI trained on my own work to give me views or feedback on what I could do better. And sometimes the suggestions could be good and they make sense sometimes. Like when I write fiction, I use AI to just give me feedback. And I think that's like a positive way to use it, to give you feedback and um what you could tweak. And yeah, I do I feel that's good, but um not like the way Johan said, not in the cannibalistic nature of AI taking away our creative agents to create stuff, but if we could use it to assist and to um show us other ways or other perspectives, I think that could be good.
SPEAKER_04:I would love it if each of you offered a closing thought. So please.
SPEAKER_03:It was really interesting to hear your your perspectives and uh hope we can keep the discussion going. Thanks for the lot of uh good examples that came up that I wasn't aware about and dig into those. So I look forward to trying out play for. It sounds like a lot of fun.
SPEAKER_02:AI is fed with uh what creatives already created. That's what you used to train the machine. So if you're taking away opportunities for creatives to make money, I think so. In the future, would AI be trained on what AI is already produced, or if like creatives are not like having opportunities due to like the cannibalistic nature of the way of prompting and the rest. So I just feel that if we could reach like a middle ground between original work and generated work, I think that would be better for the future.
SPEAKER_00:And definitely coming to uncharted territories as we go and and uh lots of questions being raised every day, and lots of new data points coming, and uh we have to keep this discussion all going on and uh and uh continue.
SPEAKER_04:Emily, any thoughts?
SPEAKER_01:Oh, I think the main takeaway is that uh as Ahmed said, it's more questions than answers at the moment, but I think I guess on a positive note, uh and it's a bit trite, but I think things will kind of work themselves out in the end, just like you know, new technologies like photography, and certainly in the kind of in the narrow sort of IP world, I think things will work out, and there'll be kind of licensing schemes, and um in time things will work out. That said, you know, AI more generally, um there are so many kind of questions, and I think the answers to those questions will very much sort of shape the world we live in and we boss on to future generations.
SPEAKER_04:There will be links in the show notes to learn more. If you want to be able to be much pretty much time, this is the thank you so much for listening.