AIxDESIGN Bookmark Library

AIxDESIGN Bookmark Library

1913 bookmarks
Newest
The Curse of Recursion: Training on Generated Data Makes Models Forget
The Curse of Recursion: Training on Generated Data Makes Models Forget
Stable Diffusion revolutionised image creation from descriptive text. GPT-2, GPT-3(.5) and GPT-4 demonstrated astonishing performance across a variety of language tasks. ChatGPT introduced such language models to the general public. It is now clear that large language models (LLMs) are here to stay, and will bring about drastic change in the whole ecosystem of online text and images. In this paper we consider what the future might hold. What will happen to GPT-{n} once LLMs contribute much of the language found online? We find that use of model-generated content in training causes irreversible defects in the resulting models, where tails of the original content distribution disappear. We refer to this effect as Model Collapse and show that it can occur in Variational Autoencoders, Gaussian Mixture Models and LLMs. We build theoretical intuition behind the phenomenon and portray its ubiquity amongst all learned generative models. We demonstrate that it has to be taken seriously if we are to sustain the benefits of training from large-scale data scraped from the web. Indeed, the value of data collected about genuine human interactions with systems will be increasingly valuable in the presence of content generated by LLMs in data crawled from the Internet.
The Curse of Recursion: Training on Generated Data Makes Models Forget
Pi, your personal AI
Pi, your personal AI
Hi, I'm Pi. I'm your personal AI, designed to be supportive, smart, and there for you anytime. Ask me for advice, for answers, or let's talk about whatever's on your mind.
Pi, your personal AI
AI Is a Lot of Work
AI Is a Lot of Work
How many humans does it take to make tech seem human? Millions.
AI Is a Lot of Work
Design and AI with Nadia Piet - Machine Ethics Podcast
Design and AI with Nadia Piet - Machine Ethics Podcast
This episode Nadia and I chat about how design can co-create AI, what the role of designers are in AI services? post-deployment design, narratives in AI development and AI ideologues, anthropocentric AI, augmented creativity, new AI perspectives, situated intelligences and more...
Design and AI with Nadia Piet - Machine Ethics Podcast
Algorithmic Filmmaking Async Class (August 7 thru September 11, 2023) | BustBright - Machine Learning Art
Algorithmic Filmmaking Async Class (August 7 thru September 11, 2023) | BustBright - Machine Learning Art
Please note: unlike previous courses, this entirely class will be done async. You will get access to a Slack channel and weekly recordings. It's a great option for anyone across the globe.Algorithmic Filmmaking is a course for filmmakers, creative coders, and anyone else interested in tools for non-traditional editing and filmmaking. Every week you’ll receive 2 hours of recorded tutorials and discussions to help you create films using machine learning, algorithms and other cutting-edge tools. You’ll also receive supplemental lectures, films to watch, and other materials.Wondering the kind of things you’ll learn? Check out this overview of my project Scream Scenes. We’ll cover these techniques and much more.Prerequisitesknowledge of coding principles is very helpful, but not requiredaccess to a Windows, Mac, or Linux machine that can run Zoomexperience or familiarity with command line is very helpufl, and comfort reading code (any code will be provided to you)Access to free version of Google Colab and Google Drive Example syllabus (subject to change)Week 1 How to create a database of video clipsWeek 2 Categorizing shot typesWeek 3 Algorithmic fancamsWeek 4 Matchcuts and other uses of OpenposeWeek 5 Algorithmic sequencing Week 6 Audioreactive and other guidance techniquesWeek 7 Free Space/TBDTBD Student Demo DaySchedule and StructureThis course will be run asynchronously. Every week you’ll receive 2 hours of recorded tutorials and discussions. In addition to the seven 2-hour recordings, there will be two open 1hr sessions with Derrick every week to answer questions and get help on projects. We will be using Slack for asynchronous messaging and discussion. Additional materials outside of the lectures and demos will be provided as video links, webpages or PDFs.About the instructorsDerrick Schultz is a designer, filmmaker and artist working with machine learning. His work has been featured at machine learning conference CVPR and NeurIPS, and has been commissioned by the New York Times, HP, and Acne. He has taught numerous machine learning courses, and has teaching experience at ITP, Parsons, and SVA. You can see more of his work on his website.
Algorithmic Filmmaking Async Class (August 7 thru September 11, 2023) | BustBright - Machine Learning Art
Semi-Conductor
Semi-Conductor
An AI experiment to conduct an orchestra inside your web browser.
Semi-Conductor
Creative Lab 5 Programme Google
Creative Lab 5 Programme Google
This year Google Creative Lab is inviting multidisciplinary creatives to apply for a paid gig (£300 per day) and collaboration up to 12 months at Google Creative Lab, London. As a Fiver you will work on projects big, small and first of their kind within Google. You will be paired with a mentor and work closely with other creatives. A Fiver is an essential part of Google Creative Lab and it’s the way a lot of our current leaders started.
This year Google Creative Lab is inviting multidisciplinary creatives to apply for a paid gig (£300 per day) and collaboration up to 12 months at Google Creative Lab, London. As a Fiver you will work on projects big, small and first of their kind within Google. You will be paired with a mentor and work closely with other creatives. A Fiver is an essential part of Google Creative Lab and it’s the way a lot of our current leaders started.
Creative Lab 5 Programme Google
Why ‘good’ AI systems aren’t actually good for anyone
Why ‘good’ AI systems aren’t actually good for anyone
“The technologies built by a few techies from Silicon Valley run everything and none of us know how they work.” - David Middleback, Re-inventing Education for the Digital Age A few months ago I was struck with the genius idea of embarking on a PhD program. My project aimed at figuring out how we can make some of those Silicon Valley AI technologies ‘better’ for everyone affected by them. It sounded simple enough, right? Well, it turns out that this simple description I was using to explain to fr
Why ‘good’ AI systems aren’t actually good for anyone
Call for Research Assistant: Explorer Database
Call for Research Assistant: Explorer Database
Tactical Tech is looking for a research assistant to enhance the information in our Influence Industry Explorer, an open-access database that compiles information about companies working with political parties and candidates on data-driven and digital campaigns.
Call for Research Assistant: Explorer Database
Call for Instructional Design Consultant: Adapting research into workshops
Call for Instructional Design Consultant: Adapting research into workshops
Tactical Tech is looking for a professional Instructional Design consultant to develop workshop curricula for the Influence Industry Project to accompany each of the learning modules into training materials in the existing format.
Call for Instructional Design Consultant: Adapting research into workshops
Product Designer - job at One Project
Product Designer - job at One Project
One Project is looking for a Product Designer to collaborate on prototypes and products that support alternative economic and governance systems. We are seeking an experienced Product Designer fluent in participatory practices, grounded in principles of equity and justice, and curious about how a world beyond capitalism would look, feel, and function.
One Project is looking for a Product Designer to collaborate on prototypes and products that support alternative economic and governance systems. We are seeking an experienced Product Designer fluent in participatory practices, grounded in principles of equity and justice, and curious about how a world beyond capitalism would look, feel, and function.
Product Designer - job at One Project
Ana Min Wein (Where am I From)?
Ana Min Wein (Where am I From)?
The work reflects on the creative potential, challenges and limitations of human-AI relationships. Available to watch on Somerset House's online space Channel.
Ana Min Wein (Where am I From)?
Channel
Channel
Three new film commissions created by artists and researchers Nouf Aljowaysir, Juan Covelli and Chris Zhongtian Yuan available on Somerset House's online space Channel.
here
Channel
AugE Next Open Call - Augmented Europe
AugE Next Open Call - Augmented Europe
CALL FOR AUGE NEXT Augmented Europe FellowshipsDigital Artists-in-Residency Programme With the support of the European Union’s Creative Europe programme, AUGE NEXT cooperation project launches four Fellowship Programmes for Digital Artists-in-Residency organized by the project partners in four European countries: Italy, Latvia, Germany, and Greece.  Each of the four AUGE NEXT Fellowship Programmes consists of a […]
AugE Next Open Call - Augmented Europe
Glaze: Protecting Artists from Style Mimicry
Glaze: Protecting Artists from Style Mimicry
We are an academic research group of PhD students and CS professors interested in protecting Internet users from invasive uses of machine learning.
We are an academic research group of PhD students and CS professors interested in protecting Internet users from invasive uses of machine learning.
Glaze: Protecting Artists from Style Mimicry
Decolonial AI: Decolonial Theory as Sociotechnical Foresight in Artificial Intelligence
Decolonial AI: Decolonial Theory as Sociotechnical Foresight in Artificial Intelligence
This paper explores the important role of critical science, and in particular of post-colonial and decolonial theories, in understanding and shaping the ongoing advances in artificial intelligence. Artificial Intelligence (AI) is viewed as amongst the technological advances that will reshape modern societies and their relations. Whilst the design and deployment of systems that continually adapt holds the promise of far-reaching positive change, they simultaneously pose significant risks, especially to already vulnerable peoples. Values and power are central to this discussion. Decolonial theories use historical hindsight to explain patterns of power that shape our intellectual, political, economic, and social world. By embedding a decolonial critical approach within its technical practice, AI communities can develop foresight and tactics that can better align research and technology development with established ethical principles, centring vulnerable peoples who continue to bear the brunt of negative impacts of innovation and scientific progress. We highlight problematic applications that are instances of coloniality, and using a decolonial lens, submit three tactics that can form a decolonial field of artificial intelligence: creating a critical technical practice of AI, seeking reverse tutelage and reverse pedagogies, and the renewal of affective and political communities. The years ahead will usher in a wave of new scientific breakthroughs and technologies driven by AI research, making it incumbent upon AI communities to strengthen the social contract through ethical foresight and the multiplicity of intellectual perspectives available to us; ultimately supporting future technologies that enable greater well-being, with the goal of beneficence and justice for all.
Decolonial AI: Decolonial Theory as Sociotechnical Foresight in Artificial Intelligence
Positive AI: Key Challenges for Designing Wellbeing-aligned Artificial Intelligence
Positive AI: Key Challenges for Designing Wellbeing-aligned Artificial Intelligence
Artificial Intelligence (AI) is transforming the world as we know it, implying that it is up to the current generation to use the technology for ''good.'' We argue that making good use of AI constitutes aligning it with the wellbeing of conscious creatures. However, designing wellbeing-aligned AI systems is difficult. In this article, we investigate a total of twelve challenges that can be categorized as related to a lack of knowledge (how to contextualize, operationalize, optimize, and design AI for wellbeing), and lack of motivation (designing AI for wellbeing is seen as risky and unrewarding). Our discussion can be summarized into three key takeaways: 1) our understanding of the impact of systems on wellbeing should be advanced, 2) systems should be designed to promote and sustain wellbeing intentionally, and 3), above all, Positive AI starts with believing that we can change the world for the better and that it is profitable.
Positive AI: Key Challenges for Designing Wellbeing-aligned Artificial Intelligence
Design4AI Symposium 2023
Design4AI Symposium 2023
We invite you to the 2023 Design & AI Symposium where we are asking: How are designers shaping possible futures? The symposium will be held on the campus of TU Delft and is presented by the Industrial Design Engineering Faculty. At this event, you'll be able to join fellow designers and businesses to explore the intersection of design and the future of AI.
Design4AI Symposium 2023
Abeba Birhane (@Abebab) / Twitter
Abeba Birhane (@Abebab) / Twitter
Senior Fellow in Trustworthy AI @Mozilla | Cognitive science PhD @UCDCompSci @ucddublin | Ethiopian currently in Ireland. She/her @abebab@scholar.social
Abeba Birhane (@Abebab) / Twitter
Deb Raji (@rajiinio) / Twitter
Deb Raji (@rajiinio) / Twitter
AI accountability, audits & eval. Keen on participation & practical outcomes. Fellow @mozilla, CS PhDing @UCBerkeley. forever @AJLUnited, @hashtag_include ✝️
Deb Raji (@rajiinio) / Twitter
Instagram effect in ChatGPT — Abhishek Gupta | Responsible AI | ACI
Instagram effect in ChatGPT — Abhishek Gupta | Responsible AI | ACI
A lot is required to achieve those shareable trophies of taming ChatGPT into producing what you want: tinkering, rejected drafts, invocations of the right spells (I mean prompts!), and learning from observing Twitter threads and Reddit forums. But, those early efforts remain hidden, a kind of surviv
Instagram effect in ChatGPT — Abhishek Gupta | Responsible AI | ACI