AI Agents is a podcast representing a collaboration between the Sia Furler Institute and The Australian Institute for Machine Learning at the University of Adelaide as a part of the Art Intelligence Agency. Through conversational interviews, the podcast explores the intersections of contemporary art and artificial intelligence with a distinct focus on how this relates to human creativity. Hosted by Tim Whiffen of Whimsy Productions, this podcast represents the work of many respected contributors to the Art and AI fields. The possibilities of AI stretch beyond imagination, and yet are limited by the lack of inbuilt imaginative consciousness in machine learning. Follow the series as it distinguishes the prospects from the constraints of this modern technology. Upcoming interviews include creatives, stakeholders, and academics in the fields of philosophy, physiology, art, information technology, and music. Subscribe to the podcast on your favourite podcatcher and consider giving it a review. If you have any feedback, examples or topic suggestions please get in contact with Tim: tim@whimsyproductions.com. Hosted on Acast. See acast.com/privacy for more information.
📻 Siste episoder av AI Agents
Her er de nyeste episodene tilgjengelige via RSS-feeden:
Ep 17: Agent Alan Thompson - Do AI Write of Electric Sheep? (00:25:16)
As you read my words now, the text has been constructed with several decisions about the words, tone, and grammar. While it seems straightforward to articulate this message rationally, it is far from pure reason and is in fact a creative and unique process to write. Can you imagine how machines would choose to write to us? If you haven't already seen an AI send text messages, you might imagine a concise and precise use of words that could only mean one thing. Language doesn’t work like that, and English is incredibly messy, and we are now seeing astounding examples of written communication from AI. Dr Alan D. Thompson is an AI consultant with a particularly amazing example of machine learning in his AI named Leta, who has recently authored, illustrated, narrated and then described her own children’s book. Leta’s Adventures is simply a must-read for 2022 and sits amongst an incredible collection of exploration and examination of AI available on Alan’s popular Youtube Channel. Alan graced the Art Intelligence Agency unaccompanied, though I am sure it won’t be long before we can hear Leta join the podcast.If you're intrigued to hear more of Alan's excellent commentary on AI, follow his work on his website, or subscribe to his renowned newsletter, The Memo.This Podcast was Produced by Tim Whiffen of Whimsy Productions in Association with the Art Intelligence Agency, the Australian Institute for Machine Learning, and the Sia Furler Institute of the University of Adelaide. Support the show: https://arts.adelaide.edu.au/music/sia-furler-institute/engagement#See omnystudio.com/listener for privacy information. Hosted on Acast. See acast.com/privacy for more information.
Ep 16: Agent Ramon Amaro - Can Machines Perceive in Totality? (00:44:28)
When AI is only worth as much as the algorithms and data it is based upon, what cultural impacts and criticisms are not afforded to what can be deemed objective data? Though we have looked at AI through the disciplines of philosophy, engineering, and art, the humanities have a lot more to do with this field as we increasingly realise the cultural potential (and indeed impact) of machine learning. Ramon Amaro is perfectly situated to identify issues with AI practice now, as a Lecturer in Visual Cultures at University College London. Ramon holds a Ph.D. in Philosophy at Goldsmiths, a Masters's degree in Sociological Research from the University of Essex, and a Bachelor of Science in Mechanical Engineering from the University of Michigan, Ann Arbor. With broad expertise, Agent Ramon Amaro joins the art intelligence agency to explore ideas of Afrofuturism, problematisation of AI, and ideas of perception in machines and visual cultures.If you enjoyed this episode, you can find more of Ramon's work and purchase his book The Black Technical Object.This Podcast was Produced by Tim Whiffen of Whimsy Productions in Association with the Art Intelligence Agency, the Australian Institute for Machine Learning, and the Sia Furler Institute of the University of Adelaide. Support the show: https://arts.adelaide.edu.au/music/sia-furler-institute/engagement#See omnystudio.com/listener for privacy information. Hosted on Acast. See acast.com/privacy for more information.
Ep 15: Agent Anthony Fraiser - Leveraging AI for Creative Content (00:24:41)
Would you listen to a story made by a machine? We've discussed how AI can collaborate with humans in art, inspiring some truly great work across mediums. But what happens when creatives take their passion to commercial products, what room is there for AI in commercial innovation? It turns out, a lot! ABF creative is one of the world's first content producers to leverage AI in the creation of podcasts. Anthony Fraiser, ABF Creative CEO, joins the Art Intelligence Agency to discuss the process of using AI in storytelling and the considerations of AI in their multicultural practice.You can find more of Anthony's work and read his book on Facebook. You can also contact Anthony at anthony@abfc.coThis Podcast was Produced by Tim Whiffen of Whimsy Productions in Association with the Art Intelligence Agency, the Australian Institute for Machine Learning, and the Sia Furler Institute of the University of Adelaide.Support the show: https://arts.adelaide.edu.au/music/sia-furler-institute/engagement#See omnystudio.com/listener for privacy information. Hosted on Acast. See acast.com/privacy for more information.
Ep 14: Agent Marcus du Sautoy - Formulating a Creative Outcome (00:31:49)
It seems that many people would rather use calculators or employ the services of an accountant than practice the language of mathematics, we've got plenty more fun things to be doing than formal equations. But is this boredom surrounding mathematics due to the language, or the subject of equations? Perhaps grammar is to a poem what math is to AI art. Can we code for creative outcomes? Is there a formula for art? It sounds daunting but Marcus Du Sautoy, author of the creativity code, is an Oxford University mathematician with fascinating insights on how we can solve for bold and beautiful in AI. Marcus joins the Art Intelligence agency to discuss the art of mathematics and the mathematics in art.You can find more of Marcus's work in mathematics here, and follow his newest book on bookshelves in August.This Podcast was Produced by Tim Whiffen of Whimsy Productions in Association with the Art Intelligence Agency, the Australian Institute for Machine Learning, and the Sia Furler Institute of the University of Adelaide.Support the show: https://arts.adelaide.edu.au/music/sia-furler-institute/engagement#See omnystudio.com/listener for privacy information. Hosted on Acast. See acast.com/privacy for more information.
Ep 13: Agent Marian Mazzone - When AI Writes the History Books (00:30:13)
The study of art history brings with it questions of identifying and classifying genres, styles, and interpretations, providing an understanding of art at a sophisticated level. With the emergence of AI within artistic spaces, it has drawn contemporary art historians to consider how aesthetics play into the analytical ‘mind' of a machine processing art. With the new aesthetic, and the blending of virtual and real worlds in artistic language, we have a lot of creative potential and we're only starting to see what that looks like. Professor Marian Mazzone is the chair of art history at the college of charleston in South Carolina and an expert in computational art history. She joined the Art intelligence agency to provide insight into how we classify and analyse AI in art.You can find more of Marian's work discussing a machine's understanding of art here.This Podcast was Produced by Tim Whiffen of Whimsy Productions in Association with the Art Intelligence Agency, the Australian Institute for Machine Learning, and Sia Furler Institute of the University of Adelaide.Support the show: https://arts.adelaide.edu.au/music/sia-furler-institute/engagement#See omnystudio.com/listener for privacy information. Hosted on Acast. See acast.com/privacy for more information.
Ep 12: Agent Tim Gruchy Pt2 - Incorporating Technology into Organic Expression (00:43:00)
As this series has explored the possibilities of artistic creativity with regard to AI, a common theme arises where AI is unable to self-sufficiently create meaningful work, but with direction from artistic intentions, it can be involved in truly remarkable work. Tim Gruchy is an artist who employs technology for multi-media art in this way. But as Agent Tim Gruchy explores ideas of sensory experiences, questions arise for AI's potential in a complex space, where as far as human experience is concerned, artists understand and pose questions no one is able to answer, let alone quantify for a computer.This is a two-part discussion, please see episode 11 before starting.You can find more of Tim's work here and follow his artistic endeavours on his YouTube,This Podcast was Produced by Tim Whiffen of Whimsy Productions in Association with the Art Intelligence Agency, the Australian Institute for Machine Learning, and the Sia Furler Institute of the University of Adelaide.Support the show: https://arts.adelaide.edu.au/music/sia-furler-institute/engagement#See omnystudio.com/listener for privacy information. Hosted on Acast. See acast.com/privacy for more information.
Ep 11: Agent Tim Gruchy Pt1 - Complex Experience through the Eyes of AI (00:26:35)
As this series has explored the possibilities of artistic creativity with regard to AI, a common theme arises where AI is unable to self-sufficiently create meaningful work, but with direction from artistic intentions, it can be involved in truly remarkable work. Tim Gruchy is an artist who employs technology for multi-media art in this way. But as Agent Tim Gruchy explores ideas of sensory experiences, questions arise for AI's potential in a complex space, where as far as human experience is concerned, artists understand and pose questions no one is able to answer, let alone quantify for a computer.You can find more of Tim's work here and follow his artistic endeavours on his YouTube,This Podcast was Produced by Tim Whiffen of Whimsy Productions in Association with Art Intelligence Agency, the Australian Institute for Machine Learning, and Sia Furler Institute of the University of Adelaide.Support the show: https://arts.adelaide.edu.au/music/sia-furler-institute/engagement#See omnystudio.com/listener for privacy information. Hosted on Acast. See acast.com/privacy for more information.
Ep 10: Agent Tega Brain - Engineering Art for Ecological Needs (00:23:46)
While data is technically valuable to industry and science, it also makes an amazing subject for artistic exploration and expression. Data is a fantastic way of knowing about something but it is a limited experience of the world. That limitation is a great constraint for the creation of art but is also a fantastic motivation to point out through art. Equally, AI is a great tool to collaborate with for art creation, but using machine learning for commercial benefit requires reflection and critique available in fine art. The expansion of AI in the last couple of decades is something that Tega Brain, Industry Assistant Professor at NYU, has noticed and is exploring in her research and her artistic expression. To discuss the limitations and patterns of AI within our cultural and natural settings, Tega draws from examples of her work and her insights from her recent book.You can find more of Tega's work and discover her insights into the convergence of engineering and fine art from her book Code as Creative Medium.This Podcast was Produced by Tim Whiffen of Whimsy Productions in Association with the Art Intelligence Agency, the Australian Institute for Machine Learning, and the Sia Furler Institute of the University of Adelaide.Support the show: https://arts.adelaide.edu.au/music/sia-furler-institute/engagement#See omnystudio.com/listener for privacy information. Hosted on Acast. See acast.com/privacy for more information.
Ep 9: Agent Lev Manovich - Machines are Already More Creative Than Us (00:30:34)
Digital culture can be separate from tangible culture but the effects and integration of digital culture make it indistinguishable as we progress further into the 21st century. With constraints like video CODECs, or internet bandwidth, it is strange to think that we participate in cultures through digital means and that this represents our achievements and behaviours. We know there are AI digital influencers, and artists have incorporated AI in their art-making process, so AI is already implanted in cultural processes. But as we continue to develop and shift digital cultures, who do we recognise as the architects of aesthetics, software engineers designing tools for artists, or artists following patterns and making products? And where does Machine Learning or AI come into it? To discuss these ideas the Art Intelligence Agency is joined by Agent Lev Manovich, presidential professor at the City University of New York and leading theorist of digital culture.You can find more of Lev's work and follow him on Twitter @manovich. You can also find more interesting research at the Cultural Analytics Lab, or read Lev's latest book Cultural Analytics. This Podcast was Produced by Tim Whiffen of Whimsy Productions in Association with the Art Intelligence Agency, the Australian Institute for Machine Learning, and the Sia Furler Institute of the University of Adelaide.Support the show: https://arts.adelaide.edu.au/music/sia-furler-institute/engagement#See omnystudio.com/listener for privacy information. Hosted on Acast. See acast.com/privacy for more information.
Ep 8: Agent Ellen Broad - Artificial *Human* Intelligence (00:27:48)
AI can be seen as entirely separate or different to human intelligence, but if we consider it deeply, it is only made possible by human intelligence. Is AI just an extension of Humans then? Data policy expert and Senior Fellow at the School of Cybernetics, Ellen Broad, is certain that any faults or success behind AI really indicate human success. The influence of Ellen's book “Made by Humans: The AI Condition” made this conversation imperative to any consideration of AI in creative ventures. Ellen joins the Art Intelligence agency to demystify so AI myths and provide some insights into the history and future of her field.If you were interested in this episode, please look into Ellen's Work and follow her Twitter. You can also find her work at The School of Cybernetics or read her highly acclaimed book.This Podcast was Produced by Tim Whiffen of Whimsy Productions in Association with Art Intelligence Agency, the Australian Institute for Machine Learning, and Sia Furler Institute of the University of Adelaide.Support the show: https://arts.adelaide.edu.au/music/sia-furler-institute/engagement#See omnystudio.com/listener for privacy information. Hosted on Acast. See acast.com/privacy for more information.
Ep 7: Agent Marcus Endicott - Being Virtual in a Material World (00:21:12)
You may know virtual bots like "Siri" or "Alexa", and you may have chatted to a bot to order a pizza, but virtual beings take these foundations to the next level. Even now we imagine personalities for our virtual assistants which amount to lines of code; we give them pronouns and ask them if they love us (just for a joke). But what happens when there is a reality like that depicted in the Hollywood films “Her”, or in "Bladerunner 2049"? Virtual assistants and chatbots have integrated into our lives without much fuss, but what about the more complex 'beings' that are being developed in virtual reality? Will these be accepted so easily? What could we use them to do if they are? Agent Marcus Endicott joins the art intelligent agency to answer these questions and discuss his foundational work in the field.You can find more of Marcus' work and follow his moderated community on Facebook.This Podcast was Produced by Tim Whiffen of Whimsy Productions in Association with Art Intelligence Agency, the Australian Institute for Machine Learning, and Sia Furler Institute of the University of Adelaide.Support the show: https://arts.adelaide.edu.au/music/sia-furler-institute/engagement#See omnystudio.com/listener for privacy information. Hosted on Acast. See acast.com/privacy for more information.
Ep 6: Agent David Olney - The Merit of Boring AI (00:30:53)
So far we've shown some ways in which artists and engineers are using AI in creative ways or in creative processes and it has been a fascinating philosophical journey. In this episode, we bring Agent David Olney (as featured in episode 4) back on the podcast to discuss why AI can be 'boring' and still innovative. High reliability is a quality in contention with creativity in many ways. As a blind academic, David uses AI to get around, assess his surroundings, or even read to him and those processes have to be reliable. These programs aren't creative in conventional ways, they don't interpret art or have creative expression, but designing these programs requires thinking beyond your own limitations, and considering the needs of people with different perceptions of the world.Find David's work on his website, or follow him on Twitter.This Podcast was Produced by Tim Whiffen of Whimsy Productions in Association with the Art Intelligence Agency, the Australian Institute for Machine Learning, and the Sia Furler Institute of the University of Adelaide.Support the show: https://arts.adelaide.edu.au/music/sia-furler-institute/engagement#See omnystudio.com/listener for privacy information. Hosted on Acast. See acast.com/privacy for more information.
Ep 5: Agent Mitchell Whitelaw - Data as a Tool for Aesthetic Creations (00:19:10)
The act of using code or data and giving them aesthetic outcomes is the project of the 21st century, the information age. With all of this information and computing power at our disposal, it doesn't seem like we are far from the kinds of machine learning that can autonomously visually impress us if it were even inclined to produce such things. Dr Mitchell Whitelaw of the Australian National University is sceptical about Artificial intelligence being inclined to produce art autonomously in ways we would accept or appreciate but has a thorough understanding of ways this technology can be used to create art with human direction or collaboration. As an expert in generative art, Mitchell's work has looked into the randomness and surprise we can get from code, but it is always the human directing the data for the truly creative work. In this discussion, we drill down into the considerations of generative art, the limitations of AI art, and how representing data is a creative and artistic exercise worthy of cultural appreciation.Find Mitchell's work on his website, or follow him on Twitter.This Podcast was Produced by Tim Whiffen of Whimsy Productions in Association with the Art Intelligence Agency, the Australian Institute for Machine Learning, and Sia Furler Institute of the University of Adelaide.Support the show: https://arts.adelaide.edu.au/music/sia-furler-institute/engagement#See omnystudio.com/listener for privacy information. Hosted on Acast. See acast.com/privacy for more information.
Ep 4: Agent Saqib Shaikh - Seeing the world through AI (Featuring Agent David Olney) (00:43:28)
When we look at a painting or gaze upon a movie scene, we gather the visual information on what objects are, why something may be moving, and what that might mean. For a machine, these inputs break down to 1s and 0s to make sense of this information through algorithms and patterns. Visual systems are incredibly complex, and replicating them for those without sight is a significant technological challenge. Agent Saqib Shaikh is a software engineer at Microsoft and project lead of Seeing AI, a mobile application that narrates the world around us. To address the capability of AI for visually impaired people, Agent Saqib, co-founder of Seeing AI joins Tim and guest host Agent David Olney to discover more about the program.Learn more about Seeing AI here.Agent David Olney is a blind academic at the University of Adelaide and a Human Performance Consultant. You can follow his work here.This Podcast was Produced by Tim Whiffen of Whimsy Productions in Association with Art Intelligence Agency, the Australian Institute for Machine Learning, and Sia Furler Institute of the University of Adelaide.Support the show: https://arts.adelaide.edu.au/music/sia-furler-institute/engagement#See omnystudio.com/listener for privacy information. Hosted on Acast. See acast.com/privacy for more information.
Ep 3: Agent Jon McCormack - The Great Partnership between Human and Machine (00:27:58)
Professor Jon McCormack of Monash University joins the Art Intelligence Agency to discuss the considerations of pairing human creative endeavours with the strict logic of computer technology. Jon's understanding of art and science give him great insight into the relationship humans have with computers. With AI being incorporated into more creative tools and forming the modern zeitgeist, why should humans care? What about these new technologies appeals to creatives and engineers and what is yet to come? Jon's experience directing Sensilab make for a perfect platform to leap from as this podcast explores the nexus between creativity and machine learning.See more of Jon's work from his research profile, Portfolio, or at Sensilab.This Podcast was Produced by Tim Whiffen of Whimsy Productions in Association with the Art Intelligence Agency, the Australian Institute for Machine Learning, and the Sia Furler Institute of the University of Adelaide.Support the show: https://arts.adelaide.edu.au/music/sia-furler-institute/engagement#See omnystudio.com/listener for privacy information. Hosted on Acast. See acast.com/privacy for more information.
Ep 2: Agent Jon Opie & Agent Ben Lancer - What Makes a Creative Mind a Mind at All? (00:38:24)
Dr Jonathan Opie and Benjamin Lancer join the Art Intelligence Agency to dispense their collective expertise on how machines can resemble organic intelligence. While machines have been able to solve the traveling salesman problem for decades, mould has been able to solve it for millions of years. If humans are capable of imagination and creativity, AI has a long way to catch up with evolved brains. So are machines capable of creativity? And are humans even capable of 'new' thoughts? This episode discusses the human (organic) brain and consciousness, and how we can model and compare artificial intelligence to gain insight on artistic machines now, and in the future.See more of Jon's work from his research profile.See more of Ben's work from his research profile.This Podcast was Produced by Tim Whiffen of Whimsy Productions in Association with the Art Intelligence Agency, the Australian Institute for Machine Learning, and Sia Furler Institute of the University of Adelaide.Support the show: https://arts.adelaide.edu.au/music/sia-furler-institute/engagement#See omnystudio.com/listener for privacy information. Hosted on Acast. See acast.com/privacy for more information.
Ep 1: Agent Sam Leach - Automatic Learning Augmenting the Creative Process (00:19:46)
Sam Leach joins the podcast to discuss his artistic work in his latest collection titled Fully Automatic, which featured at Sullivan & Strumpf in Sydney in 2020. Sam excels as an artist by his own merit, but what if we pair him up with some machine learning? Dr Matthew McAuley from Belfast helped Sam implement algorithms from a DCGAN, (Deep Convolutional Generative Adversarial Network) to automate the process of studying aesthetics through time. Sam's previous award-winning work also went into the mix allowing the AI to produce images which Sam painted resulting in a truly surreal outcome. This episode discusses the realities of working with AI as an artist and the future prospects of creative augmentation using machine learning.See more of Sam's work on his Instagram, or check out his website.This Podcast was Produced by Tim Whiffen of Whimsy Productions in Association with Art Intelligence Agency, the Australian Institute for Machine Learning, and Sia Furler Institute of the University of Adelaide.Support the show: https://arts.adelaide.edu.au/music/sia-furler-institute/engagement#See omnystudio.com/listener for privacy information. Hosted on Acast. See acast.com/privacy for more information.
