Fantastic Futures 2024 - Day 2 - Session 6

This transcript was generated by NFSA Bowerbird and may contain errors.

Thanks Rebecca and thanks everyone for such wonderful conversations. So we're really pleased to be joining you for this panel that we've sort of framed as a UTS and Digital Preservation Coalition collaboration, but also to introduce you to the iReal project. So the IRL project, which I explained yesterday, is a longer title for inclusive requirements solicitation for AI in libraries to support respectful management of indigenous knowledges. So I'm going to keep saying IRL so I don't have to do that again. But what we'll be doing in this panel is introducing you to the Scoping Project, iREEL, which aims to develop a model for responsible AI systems development in libraries seeking to include knowledge from Indigenous communities and specifically drawing on experiences of Aboriginal and Torres Strait Islander communities in Australia. What we'll talk about is how the project addresses the issue that libraries and archives, as we know, both nationally and internationally, hold collections from many indigenous communities with varied approaches to collection access, engagement, control, and governance. These different approaches need to be understood and addressed by collecting institutions, by researchers, AI developers, and GLAM professionals to understand, develop, and implement appropriate requirements for the responsible care and protection of Indigenous knowledges and data sources from communities. So we've had a little bit of a change in the session. Unfortunately, Professor Jason DeSantolo, one of our colleagues, can't make it and he sends his apologies for not being here. But I have a few notes from Jason that I'll also share. So, what we've decided to do as well is you have to hear from me again for a few minutes as I go through some context of iREEL and then we're going to turn to a conversation with Lauren and Robin who will introduce themselves further to talk about the approaches and the methods of the project. And I guess I wanted to sort of contextualize this talk to say, even though Ingrid this morning quoted me talking about the work of Professor Chelsea Watergo and her views on hope, is that although we're bringing quite a critical view to the archives in the iReal project, we're really trying to explore the interventions and the methods in a really open and practical way, and that's what we hope to share with you, so that we can identify the spaces of agency for Indigenous people in these processes. So, we do have hope. We're just taking it through that critical lens. I want to acknowledge we're on Ngunnawal, Ngambri country and I also want to acknowledge other First Nations people in the room. And I guess in the context of talking about the Indigenous Data Sovereignty Movement, really invite people to participate in this dialogue and In the context of that as well, I really want to thank speakers for their contributions. I've learnt so much here in the last two days and I'm very thankful to have the opportunity to learn in what we all know is a really complex and emerging space. So, I'm going to run through a little bit about iReal, and as part of that, I want to acknowledge the team. I'm not sure if anyone's actually joining us from the UK, but just to give some context, iReal is a six-month BRAID, so Bridging Responsible AI Divide Scoping Project, led by Professor Paul Gooding at the University of Glasgow. And Paul's team includes Dr. Rosie Spooner and Dr. Abdinor Buic. Along with the team at University of Glasgow, we've been working with King's College London and the Digital Lab, and our main partner there, along with the broader team, is Samantha Callaghan. Lauren, myself, and our colleague Monica Galassi are joining from University of Technology Sydney, and Robin Wright is our representative from the Digital Preservation Coalition. And I guess when we're talking about the IRL project, we've sort of been contextualising this as a big leap into a very interdisciplinary and a collaborative project. It's been, you know, interesting working online. Lauren and I had the great fortune to attend, and I'll talk a little bit about the workshops, a workshop at the King's Digital Lab recently, but a lot of these conversations are happening not only in different cultural contexts, different spaces internationally, but also often on teams, which has brought, you know, great opportunities but also challenges in the project. But ultimately, as we'll describe today, IRL seeks to develop a model for responsible AI systems and really thinking about what are the spaces, as I mentioned, that Aboriginal and Torres Strait Islander people in the first instance might be able to seek intervention into systems development. So these are some slides from the Glasgow team so I'm just going to run through these very briefly just to point out the context again that not only do we have collections relating to Indigenous people held here in Australia but also in the UK and other international spaces and as I mentioned earlier In each of these GLAM contexts with different legislation, different workforce, you know, principles, policies, protocols, they all have their own perspective on data governance. So what we're trying to do is really start to uncover and pull apart some of them assumptions and to start thinking about that in the context of AI systems development. So in a six-month project, which, you know, it's a big ask, but we're also hoping to develop what is an actionable, pragmatic, and scalable model for Indigenous community involvement in the assessment and development of AI systems via requirement solicitation. So as I mentioned, part of our methods for engaging in the research project have been to run a series of workshops. We've had one here at the National Library of Australia in July, another in the UK, and we have a follow-up workshop happening soon online. And our outputs will be to have a symposium to share what we've learnt, really importantly as a research team, what we heard from our participants at the workshop, what we heard from community advisors and I'll bring Jason's comments in in that context. And as part of the sort of dissemination, we'll be sharing what we hope is, you know, working towards a prototype model for inclusive requirements solicitation that's led by the King's Digital Lab and the Glasgow team. And we've done an extensive literature review, which has been really exciting, and we'll report on our findings, obviously. And our partnership with the DPC really is to make sure that we can As a scoping project, get this information out to as many people in as many ways as possible. So there's a QR code because we all like QR codes. If you want to go and have a look at the King's Digital Lab page and of course that page Is it okay? Oh, it's enormous. What's the room like? And yeah, just acknowledging in that context of interdisciplinary that there are many people involved in this conversation. So, if you go to that link, you'll get a sense of the partnership. So I guess just to contextualize the core aims of the project, we want to understand or gain a deeper understanding in the project around the rights of Indigenous people to have data governance and its application, specifically in libraries. And that's been so great to be hearing the work that's happening in the GLAM sector, both here and internationally. I really enjoyed hearing the work of ACMI earlier and just thinking about how we participate together in shared learning and understanding to activate some of these solutions. We will develop methods to support people to better be informed about both the technical aspects of AI design, deployment and customization in libraries, but also think so centrally about how we bring communities, families, people into the conversation around the design and critique of AI systems and I guess for Lauren and I that's a large part of our drivers. I'm going to run through a couple of our core principles but just in terms of the aims we really want to assert that our participation in the project is thinking all the time about the right of reply and for people to be thinking about how they can participate in those design questions and their requirements solicitation as they see fit for community priorities. So, just to touch on a couple of the principles that were brought strongly to the project, and these were a focus of workshop one at the National Library. We've drawn on the really important work and acknowledging our colleagues from Aotearoa of Professor, Distinguished Professor Linda Tuiwai-Smith and decolonizing methodologies and always sort of looked to the 25 indigenous projects that are within. decolonizing methodologies, but we've really drawn on the idea of intervening, and I'll read out what the intervening method is in a moment. So thinking about how we're proactively contributing to discussion and seeking to address change and structural change in our work. And also the Indigenous Archive Collective position statement on the right of reply has been really central. So Smith's description of intervening really focuses on the role of research in being action orientated. And what we've asked of people who are participating in the workshops is to think of themselves literally as being in a process of becoming interested in involved workers for change. So to intervene actually means that people actively participate and question their processes and their assumptions when they engage in the project. So, the example by Linda Smith is the community itself invites a project in and sets out its parameters. So, I think one of the challenges in scoping research is to work to community identified research questions and we're trying to think about that in our process. But I think it's also in that context for us intervening means having those really uncomfortable conversations and again really am thankful to hear that this community is talking about that in your work and talking about the ethics of AI so consistently. The other thing that's central to us is the assumption that the archives, as we know, aren't neutral. You know, thinking about the idea of the reckoning that we need is that people need to actually understand that a lot of Indigenous people aren't aware of collections that exist that relate to them. And so the right to know is really critical. Seeing the pictures earlier on screen of photographs of community members in different international contexts begs the question, do people know that these collections exist and what's our effort and what's needed to actually make that information proactively available to communities? So we assert in the project in terms of the AI development the right of reply, but we know that the right of reply is contingent on that right to know. So we can't assume that people are just going to come and find stuff, we have to get out there and let them know. And I guess as core to this is the body of work that sits around the idea that people active participatory agents. So it isn't a community that sits to the side as that crowdsourcing kind of moment, but that people actually have fundamental rights to seek this information. So we've responded to the principles from the Indigenous Archive Collective right of reply in the project. I'm not going to go through these in detail. I think that they'll be disseminated at some point, but we're really consciously thinking in terms of the IRL project, what does the right to know look like? What does participation look like? The ongoing question of cultural safety, of course, is really important to us. I think coming from an Indigenous research institute, we're really aware that, you know, if we're dealing with complexity all the time, it's our bodies that have to be participating in this work, so health and wellbeing are also really critical. Key is also the notion of consent and in the context that we're working with historical collections, how do we embody and think about notions of free, prior, informed consent when people didn't have the ability to participate in those processes previously? The collective also asserted a couple of other things that I think have come up here as part of the conference, seeing institutions as facilitators and not owners, so working collaboratively and working through real participatory approaches. And that's not just, I guess, in an engagement sense, but thinking about people's right to be informed and also provide information back. And at the core of everything that we're doing, we know that we don't have the answers to everything. Maybe we don't have answers to a lot of things, but we want to be advocates and we actually want to participate in this dialogue. We've talked a lot about Indigenous data sovereignty here, but also in the project, and drawing on the work of Mayim Ngari Wingara as Australia's leading collective, thinking about this motto, so Indigenous data sovereignty everywhere and all the time. This has to be key to the work that we're doing. And just to give some context of the literature review that will follow, and recognizing Abdinor and Monica for their extensive work, we've had some sort of methodological and methods foundation thinking about this work. And so thinking that we can't extract AI from the body of indigenous research that has been undertaken. So there's been four or three themes here. The anti and decolonial theories and praxis in Indigenous politics. Key to that and some of the conversations we've had here around refusal and rejection and where does that sit in the context of settler sovereignty, movements like Land Back and the recognition of Indigenous people's sovereignty more broadly. Indigenous decolonizing methodologies, we've spoken about the interventions, but also thinking about this in terms of our practices and how we inform the work that we do. I think in institutions, when you're framing projects, how do you start to sort of consider your positionality and the work that you do related back to these decolonizing methodologies? And I think I mentioned this yesterday, seeing the real difference between a decolonising and an Indigenous approach, and really excited to hear some of the discussions earlier from the panel about Indigenous futurism and seeing Anne-Marlene Quay-Mullina's work here at ANU. I was really excited to see that, so tapping into that broader body of work. And just one last thing to mention before I head over and start a conversation with Lauren and Robin. We're sort of looking broadly at, in Abdinora and Monica's literature review, is looking at technology, that intersection with Indigenous data sovereignty and governance. Trying in this project to pull apart those ideas of how colonialism and those structures are embedded and replicated in AI instead of IA there. I've always been known to have a timer with my slides. And thinking again about that question of agency, so where do we take responsibility? And I think that comes back to that question of institutional leadership, personal leadership, and actually being conscious of the work that you're doing. And for us, so critically, and as we turn to the conversation, thinking about our methods. We know that data sovereignty is an assertion of rights. We know that the question of governance comes down to decision making, but it's crucial for us to think, how do we do this work? What does it look like day to day? So there's a little bit of context to IRL. What I'm going to do as I sit down and find my questions for Lauren and Robin is ask them to introduce themselves. Thanks. Is this fine? Thank you, Kirsten. Hi, everyone. My name's Lauren Booker. I'm a postdoctoral research fellow at Jumbunna Research Institute in the Indigenous Archives and Data Stewardship Hub, which is led by Kirsten Thorpe alongside our colleague, Monica Galassi. My family on my mother's side is Garigal, so north of Sydney, connections to northwest of Sydney. And through my father's side, my father is from Nagasaki Prefecture in Japan. I'll say a little bit about the work that I do currently and have done. So my focus and kind of what I guess I bring, context of what I bring to this project. I'm a CI on the IRL project. I engage in work that is focused on reconnecting collections of Indigenous cultural intellectual property, Indigenous knowledges, records, reconnecting those collections to communities. and supporting communities, getting their materials out of GLAM institutions. I also do work in the support of the organization and management of community archives. And more recently in the last few years, alongside Kirsten, we've been doing a lot of, I've been doing a lot of work looking at supporting better access to records for stolen generation survivors and their descendants. But my background is also in archival support and collections support on language revitalization projects, and also setting up digitization workflows for photographic collections, which I very happily continue to do on an occasional basis because it's fun. But my personal research, which aligns, which is one of the reasons why I was quite interested in this project and wanted to get on board, was I look at the histories and the legacies of 19th and 20th century race science and the collections that have been left over and still used, depending on what they are, to this day that are held in in libraries, archives and museums. I'm particularly interested in those that are held in an international jurisdiction, how that legislation policy kind of environment maps over to our own, particularly in a research ethics perspective. We have a very particular kind of research ethics Here that we go through to use certain collections that either contain incredibly personal information and body surveillance, we heard a bit about that earlier, records of people's families, but also to the ancestors as well, and genomic material. I've been very interested in the intersection of that work, particularly because I look at correspondence papers in large collections, people's personal papers, and thinking about the intersection of that work and the work of auditing and the work of going through materials such as that. and machine learning, AI, and both the, I guess, I don't wanna say pros and cons to be so black and white, but the risks and the opportunities of both of those things. So yeah, that's a bit about me. Robin, do you wanna introduce yourself? Yes, thank you, Lauren. My name is Robyn Wright. I'm a non-Indigenous person, but I'm very happy to be working on this project. I'm the head Australasia and Asia-Pacific for the Digital Preservation Coalition. For those of you who don't know, the Digital Preservation Coalition is a global community working together to bring about a sustainable future for our digital assets. It was established 22 years ago in the UK, and it's a membership organisation made up of institutional members, many memory organisations, of course, archives, galleries, libraries, but also commercial and scientific and other types of organisations that have significant digital collections. We have over 160 members worldwide and the DPC established an office in Australasia and Asia Pacific in Melbourne in 2020. So we now have 24 local members here in our region and an active community who are working together on the preservation of digital assets in our region. And yes, many of the people here today, we're very pleased to say, are members of the DPC, and the NFSA itself is a member of the DPC, which is great. The DPC sees an important role for the Australasia and Asia Pacific Office in helping disseminate information about indigenous data sovereignty and the respectful management of indigenous knowledges in digital collections as part of our activities in our region and for our members around the world. So we want to work with and support local indigenous individuals and communities with their digital preservation needs. and to help disseminate the requirements for the appropriate management of Indigenous collections and knowledges across our global membership. We've started by speaking with a number of Indigenous practitioners and researchers and supporting projects that are aiming to improve the digital preservation of Indigenous content and knowledge. And we recently updated one of our own key resources, the DPC's Rapid Assessment Model, or RAM, that some of you might have used, to incorporate some of the requirements of the Tandania Adelaide Declaration. And we're also very pleased last year to be able to make a connection between Dr Paul Gooding from the Library Studies and Digital Scholarship Department at the University of Glasgow, where the DPC is based, with Dr Kirsten Thorpe and Dr Lauren Booker from the Jumbunna Institute for Indigenous Education and Research at UTS, which we think is a fantastic connection. and they were able to explore opportunities to fund research around enabling responsible AI when working with indigenous content in the library and GLAMP sectors. So this has led to the iREEL project and it's been great to see the project develop and to receive funding and establish the partnerships necessary to explore AI in libraries and GLAMP organizations and how they're developing the responsible management around indigenous knowledges. So the DPC is very excited to be part of this project. I attended the first workshop that was held here in Canberra earlier this year. And DPC staff in the UK have attended the workshops and events held over there in the King's Digital Lab recently, which is also great. So we hope that the DPC's Australasia and Asia-Pacific Office will play an important role in disseminating and amplifying the findings of the project to our members around the world. Our members are obviously interested in the responsible use of AI for managing and preserving their digital collections. And of course, many of those collections around the world, as Kirsten said, do contain Indigenous material. So we're very happy to be a partner on this project and hope to be involved in promoting and implementing the findings with our members, both here in our region and around the world. So thanks, Kirsten. Thanks Robert and Lauren. One of the things that I wanted to sort of cut straight to in the iReal project is I guess the scale of the question compared to a six month scoping project and what I've tried to contextualise is some of the principles and methods that were brought to the fore in the project to sort of ground us. But I wonder if you have any reflections, if I start with you, Lauren, on particularly, obviously, we're talking about use of Indigenous data in AI systems development. And one of the questions that has come up to us, or to us throughout the project, has been what data are we using? And I wonder if you are able to reflect on some of those experiences of your own personal responsibility, thinking how you engage in that space yourself ethically to suggest in a six-month project how you might imagine community participation and partnership. Six months is very short, but that sentence should not be a reason to not do something when it comes to engagement with community, or at least not to put certain things in place. I think with the project that, with the IRL project, we were looking at engaging through, which we'll talk about in a little bit about our methodology of yarning as methodology, Aboriginal research methodology, which isn't just a conversational method, it's something more than that. But through that, it's a relational methodology, and so through that, we were able to bring, through our networks, bring people onto the project through and with the project through the way that we structured it around workshops and we're able to have those conversations with Aboriginal and Torres Strait Islander experts within the field, the field that we're contextualising, so particularly libraries but GLAM more broadly, the field that we're contextualising the project in. And, you know, that needed a lot of planning very, very early on. And so I think, you know, if you're, when you're going into, I mean, any project doesn't matter if it's, you know, like a four-year ARC, big bucks or a scoping project, as early as possible, you know, having First Nations CIs on the project, if possible, should be driven by people who are in the leadership team. But also thinking about putting those mechanisms in place as early as possible. And recognizing that if you have a short time, you know, trying to, and being transparent about that, I think, as well, being transparent about the time constraints. with the people that you want to engage with so that every that's all up front in the beginning but yeah it's it's it's a really um i mean scoping is so important short projects are also that can be brilliant because you know it's like that It's exciting and it's quick and you wanna get all this stuff. And someone said yesterday, we're not just talking about AI, we're doing, and which is awesome, but what we really are bringing to our project is we are talking about the project and that for us is doing. I think having those conversations and engaging in working through those sticky, I mean, they're wicked problems, a lot of these things we're dealing with. They're not gonna be solved. No one's solving them in six months. But being really active in that communication, I think, has been what we're... With the people that we have on the project, so also, you know, of course, with Jason and Gaydrian, and the project team more broadly, I think. Yeah. I might jump to you with that question in a moment. Robin, as a bit of a follow-up, to maybe describe the participation in workshop one and one of the things that we took to that was obviously that focus around intervention but also I guess bringing a toolkit for people to participate in a dialogue. I loved hearing Acme talking about the dialogue and the discussions that were taking place, but feel that people could participate in some of that unravelling. You know, looking at this question of if people don't know what collections exist relating to them already, how are we then going to introduce the question of AI into that space in whatever kind of range of AI tools that we're talking about? And I wondered, I know Jason, so just to briefly say, Jason, I've got a couple of things that I'm gonna weave through with these questions. We've had Jason as an advisor to the project, and I'm really thankful that Jason has brought a community liaison officer into the discussions, Gayjoon Hussan, who, they've had sort of five key things that they've talked about. And they participated in workshop one through a presentation, and that was thinking about self-determination and knowledge sharing, thinking about transformative partnerships and outcomes, so not just a knowledge, a question of knowledge, ownership, and consent, but questions around exploitation in knowledge production of labor and reciprocity, what does that look like. They also focused on the repatriation agenda and the methods for crafting stories that align those values, kind of hooked back into Indigenous research methodologies, but taking us back all the time to the local and what does this mean for people in a community context. and particularly thinking about those ideas of people being in crisis in community. And then that broader focus that I guess, you know, GLAM has real opportunities of looking at cultural memory, cultural resurgence and working together for public education. And I wondered in the context of Jason asserting those themes and Gaydrian in the workshop, what was your takeaways of participating in workshop one at the National Library? Ah, well, I think the thing that struck me most was the complexity of the data gathering process. When people talk about AI, there's so much suggestion that, OK, we'll go and get the data. And what the workshop taught me very much was the need to sit back and listen to the complexity of the situation and complexity of the data gathering process and the people who were using the data and were the subjects of the data and how there was a real need to slow down and to stop and to think and to reassess different approaches to what the data even was, let alone how it was going to be used. It was a great experience for me to think more deeply and in more complexity about the things that AI was doing and the type of data that we were trying to address in the project. Yeah, I think that just jogged my memory about something. Thinking about kind of the responsibilities that, you know, Kirsten and I talked a lot, continue to talk a lot about, you know, in our projects, the responsibilities that we have carrying forward with certain projects and engaging with. So for this one, for example, you know, looking at what kind of, what data set were we going to potentially look at in this project? And yes, there are open data sets, they're open, they're government, they're up there, people are using them already, blah, blah, blah, blah. But then it's that, in a way, that doesn't nullify our response. My responsibility, because I'm speaking for the both of us, but our responsibility in thinking about what are the repercussions of us pointing to that data set of us engaging with that data set, whether we're the RSCs, which we're not, whether we're doing the behind-the-scenes kind of tech stuff, or if we're introducing that. So I think those kinds of responsibilities we carry throughout all of our projects, and is something that we deeply spend a lot of time considering, and it doesn't map over to, a lot of the time, sometimes it does, but not all the time does it map over into institutional timeframes, particularly in an interdisciplinary space. I think sometimes, you know, the space that we work in, I think when you work within your own sector, you've got those expectations of, you know how long usually things take to do, to converse, that kind of thing. And we do need to figure certain things out before we can move forward. And it works slightly differently to other disciplines or other research spaces. So I think that was really forefront of my mind and definitely forefront in a lot of people's minds that we're speaking about in the workshop because we had a lot of other people in the workshop, a lot of Aboriginal and Torres Strait Islander library professionals and researchers and scholars who similarly have those responsibilities and think about those things a lot. So there was lots of conversations about that multi-layered level of responsibility. That's not just, even if you put in an ethics application, it's not just that. There's so many conversations that need to be had or things that need to be considered. Yeah. So it's a really interesting point in terms of I know yesterday I heard people talk about slow AI and in the workshop at the National Library a colleague in the room said a statement that was something that came up like just chill and it was kind of interesting because in the translation of that work I guess we were sort of also addressing cultural understandings of time and space and slowing down versus I guess the sense of fear and inertia of people sort of starting to realize that if we are to have the reckoning in GLAM and we are to look at the responsible use of AI and to have Aboriginal and Torres Strait Islander people participate in decision making about what are the requirements that we want these systems to use, that there was some tensions around slow and fast and the urgency compared to the need to be respectful. And I wonder, in terms of the methods and approaches, because I think that's something that we've really asserted in this project, that we're thinking and contemplating, have those framing of yarning and the research methodologies assisted us in guiding the project in the right way? Either of you. Well, maybe I'll start just talking a little bit about yarning as methodology, because we've had feedback in different circumstances about yarning as being understood. It's just a focus group. Isn't this just a focus group? Which, and in one level of using it, yes, it does seem like it's just a, it is a conversational method, methodology, but it's much, much more than that. It looks at, it encapsulates the understanding of relationality amongst research participants and researchers and this, you know, I'm sure we're all quite well versed in ethics and the notion of prior relationships conflict of interest and things like that. It doesn't therefore mean that there is none of that, but it's recognizing that prior relationships, it adds another layer of accountability. It adds what I was just speaking about before, this accountability that we have. In yarning methodology it recognises that. It also recognises that conversations are multi-layered, so there's lots of things going on. It's not just we're talking in a room and people are interjecting and there's a sense of excitement or there's a sense of annoyance. There's all of these different kinds of conversations that are happening. And it's, yeah, it mirrors relationality. It mirrors how we know each other or how we don't, or, and it doesn't, and I think in workshop one, it was, you know, we built the workshop through our, we developed the workshop, sorry, through our networks, and it was also some of the project team from Glasgow were able to come. Sadly, our KDL colleagues could not come, but Sam joined us until about 4 a.m. her time on Zoom, which was heroic. But we, because of the way that that workshop was set up, we were able to engage with this methodology. And so I don't think that, you know, it's not something that will always work in that way. Like it needs to be indigenous-led. It needs to be relevant. You need to be using relevant methodologies for what you're doing. You can't just tack something on. So you have to think about that quite deeply. Yeah, that's my thoughts on that. I was just going to say that I saw it very much as a process that opened up a space that I hadn't expected. I personally hadn't expected, an unexpected space. And you tend to go into these research projects saying, we've got six months, we've got to achieve this, and this is the day that we're going to do the workshop. And what the methodology did is it obviously allowed space for a number of voices, but it also allowed the discussion and the ideas to be led in a whole lot of different ways that I hadn't anticipated and obviously other people hadn't anticipated either. And I really appreciated being part of that and seeing those spaces open up. I'm not sure what the end result of that will be, but I really thought that was a fantastic part of participating in the workshop. Yeah, I think methodology like that can, it shows how interconnected things are. We don't do our work, like as academics, or any work, you don't do it in a vacuum. And so when you're talking about one thing, it then opens up into a completely other space. And sometimes, I mean, even sometimes you'll be sitting there like, oh, wow. But then there's always an element that there's a reason why people bring things up. There's a reason why we meander down a path and end up talking about some seemingly random thing that happened about 10 years ago. There's always a story there and reasons for people to have those conversations. And I think that goes back, you were speaking yesterday, Kirsten, about how some people came to the workshop, quite a few people came to the workshop and up front were like, oh, I really don't know much about AI. I don't really know how useful I'm going to be as a participant. And it allowed for that space for people to air those things and then also to just speak about We were speaking about machine learning and AI and the issues and the risks and the opportunities. We were speaking about it all day even when we were going through periods of time of not directly speaking about it. So it's recognizing that everyone had something to add in that conversation and it allowed for that. So yesterday we've heard the theme of hope and the fantastic futures. I guess in looking at a six-month scoping project, do you have any ideas in terms of, you know, there's a community of people here and online? And Robin, thinking about the network in the DPC, what a project like iReal can bring to the fore? And I guess thinking about the need for toolkits and the ability for people to have some things that they can tangibly take into their practice to say, you know, maybe this model will work, like the NFSA has talked about having a set of principles to guide their work. Do you have any sort of aspirations from the iReal project that you see can be the pragmatic, the actionable as outcomes? Obviously, we're very excited to be able to see Indigenous voices in the discussion that we're taking out to the global community. We think that's a really important part of having the office here in Australasia and Asia-Pacific is that we can work with those voices and amplify them around the world. I think that's the main message from us. I think it's made me realize there's a lot of work that needs to be done around knowledge translation. Yesterday, people spoke to that. Today, everyone's been speaking to that. That's an unfinished space. And it's in the work that we do. There's a lot of expectation on and assumptions, actually, on First Nations people to do knowledge translation all the time, always explaining how things work, why things are significant. We have to say why our ancestors are significant to have them returned. Like, it's why our records of our grandparents, our parents, or ourselves are significant to have things. There's expectations of those kinds of translations there. I think there's a lot of opportunity to start to build some, to start to skillshare, to translate some of these ideas around. translation is probably not even the right word. I think it's like opening things up. And lots of people have been doing that, I think, in the last two days. It's been awesome seeing. And I'm not a heavy technologist. I do say, like, I really love, like, metadata, blah, blah, blah. But, like, still people will be taught. Yeah, I couldn't follow. But it was, I think people have done a really, really beautiful job of doing that. And there's so much, like, so much advocacy work or just making sure that we need to start doing that more often and that's an expectation. that if you can't explain something, if something can't be explained clearly, or like how is there gonna be mutual benefit if that's something that can't be described? And I think that the iReal project has really kind of opened my eyes into a lot of research spaces that I was previously unaware of. And it's exciting though, it's like it's a really, because so much amazing stuff can, obviously the last few days have shown so many amazing things that can happen with AI. Thanks Lauren and Robin. Just as one sort of final point to touch on, going back to Jason's reflections about communities in crisis. At workshop one, Gaydrian Hussan joined us talking about this tension around what he had perceived as living culture being in community and transmitted through family and through connections with land. and country and then talking about the idea that glam represents what he termed as being quite dead collections because people weren't engaging with them despite obviously in recognizing you know the incredible work that institutions are doing to make sure that that isn't the case and one of the things that really came to the fore for me in that workshop was Gadrian talking about the current politics of the mining interests in his community in the Northern Territory, the surveillance of the state in terms of, or the Territory in that case, in terms of access to schooling and the need for roads to be repaired. And all of a sudden we found ourselves in a room talking in quite a privileged space about, you know, a UK-funded project about responsible AI and even thinking about the tensions of, you know, are we actually talking about Indigenous knowledges or are we talking about collections that document Aboriginal people through surveillance? And when Gadrian talked about the potholes, it brought us to this question that goes back to Jason's comments of, you know, how do we all recognise the privileged spaces that we work in and start to support communities in crisis? And I don't know that there's a link that we can necessarily find yet in terms of the opportunities of AI, but thinking about those extractive processes and the demand, the ecological damage of AI, how that thought resonated with you, with Jason and Gadrian coming in as community spokespeople, thinking about, I guess, the responsibilities of GLAM, but also the work we're doing linking back to community empowerment and self-determination. I was going to say, I think that's a real challenge to the DPC and it's something we're thinking about incorporating into our own processes, but that it will be an ongoing challenge that we want to take on board as part of the global organisation. I think having Jason and Gajarin come in and speak to the room and having that having that kind of interjection of their thoughts on the project, but also on the use of machine learning AI. I think it was going back to what I was saying before about nothing happening in a vacuum and seeing the interconnectedness of everything. and approaching things critically as well and understanding that we're not, yeah, that we're not gonna have an answer in a six month project, but seeing how it's so, the idea of, everything is so interconnected, also the idea of labor as well, like asking people to be on projects that aren't priorities, like what is a priority for the communities that you work? And you know, there isn't one, so there was concerns, some of the risks that were talked about, in the workshop, we had long conversations about risks and a lot of people were like, I think there's a lot of assumptions about what the risks are. And then someone else was saying, well, you know, the biggest risk for me is the homogenization of all of our communities. This idea that it's, you know, it's everyone's the same and there's one big indigenous community, one big Aboriginal, you know, like that's, it's such a risk. And thinking about, you know, Thinking about the relevancy of the work that you're doing to the people that you want to have involved. Who's instigating the research questions, I think is something that I come back to a lot. Yeah, whose research questions are they? And how do they actually, like, who are they? Yeah, who is it benefiting? Is it benefiting the institution? And if they are, that's fine. But it's being upfront about that. I think that there are projects that benefit institutions because institutions want to do projects. And so, but it's about saying that and being really clear about that and then using and then aligning your methodology, aligning your governance structures to those things. I think that kind of transparency is really important and I think that's what Jason and Gaydrian's presence made me kind of jogged in my mind. Yeah, thank you. I think Lauren talking about them as wicked problems they are, the real complexity of the moment that we're in. And I guess for everyone here, we're really looking forward to disseminating the research. Lauren, I don't know if you wanted to share in terms of the community. we are thinking about our responsibilities of data sovereignty isn't just about thinking about the data but also you know how do we disseminate research or how do we disseminate projects in ways that are packaged up differently to different communities so I don't know if you want to talk a little bit about the artwork that Yeah, so for one of the outcomes of the project, at least for workshop one, the project is still very much ongoing and the project partners, in recognition of the interdisciplinary and international nature of this project and the 4 a.m. 10 p.m. time split, We're focused, of course, on workshop one and so we've taken some initial findings. We see these workshops as also, and this is obviously openly transparent within the research participants, that it's an opportunity as well to produce data which we're going to analyse. So the outcomes of that have been shared with an Aboriginal-led design company, and we've asked them to look at the visual communication side of some of the conversations that we were having, so looking at what GLAM professionals, both indigenous and non-indigenous, are thinking at the moment around the workforce, around opportunities, around risks. So we hope, yeah, that'll be soon. We'll get some of that VISCOM stuff back and then that'll form the package of some of the findings that we're gonna put out there. So, thank you so much everyone for listening to us, listening to me again. Thanks for the opportunity to share the IRL project. We're really excited for that to, the results to be disseminated. Thanks to Robin, Jason and yeah, the wider team at Glasgow for participating in this. We have a poem that we're going to share but we might leave it if you want to speak to us outside. Thanks everyone. Thank you.