Community Research
Community experiments with artists interested in playful experimentation, p2p education, consensual tech, and DIY cultural production with AI. Stewarded by Emily Martinez and Sarah Ciston.
Who is this by/for?
We are artists interested in working with texts and AI using our own archives and “small data”. We want to explore ways of practicing care and support through our technological ecosystems. We want to deepen how we bring queer, feminist, and decolonial methodologies into our everyday lives and work. We want to experiment with glitch, error, and infecting the larger AI systems with a plurality of stories as a strategy of resistance. We are curious to learn if/how these approaches can help us find openings, write new narratives, and create spaces that can hold us in ways that these systems were not designed to.
Our Goals
- Making AI more accessible and pleasurable for beginners in a supportive environment.
- Centering historically excluded and misrepresented people and knowledge, respectfully and consensually.
- Developing deeper practice with queer, feminist, and decolonial methodologies.
- Working with language models and generative AI
- Working with “no code” tools that lower barrier to entry and allow us to prototype and test ideas with ease
- Setting up guidelines for preparing datasets that respect the contributors and protect their data sources.
- Making machine learning models from “scratch” and fine-tuning LLMs with our custom datasets
- Creating publications and toolkits
Our Work
🦜 p5.js Critical AI Tutorials, Google Season of Docs, 2024
Team
- Sarah Ciston Technical Writer
- Emily Martinez Technical Writer Advisor
- Minne Atairu Technical Writer Advisor
Tutorials
- 1 Chatting with/about Code - Using generative AI to code with p5.js
- 2 Critical AI Prompt Battle - Build a tool to compare generative AI prompts
- 3 Critical AI Sentiment Dataset Explorer - Examine the datasets that run AI tools
- 4 Critical AI No-AI Chatbot - Make an interactive chatbot without using any AI
🐚 Imagining Better Large Language Models, Workshop, Mozilla Fest 2023
This workshop peeks under the hood of large language models like GPT-3 and ChatGPT, in order to understand what actually goes into the datasets that shape them. Are these the automated systems we want? Whose voices, visions, and stories are captured in them and whose are excluded, harmed, or undermined? Join us to start building the text datasets and language tools you want to see in the world.
🌱 Love Corpora, TBD
Love Corpora will be a collection of datasources featuring texts that orient towards life and liberation; created with community input and contributor consent; metadata will provide full transparency about contents, licensing terms, and collection methodologies.
About Us
Emily Martinez is a new media artist, 1st generation immigrant/refugee (Cuba > Miami), and a self-taught coder who believes in the tactical misuse of technology. She started Unsupervised Pleasures as part of her ML5.js Fellowship project to document some of her research and experiments with LLMs.
Sarah Ciston is a self-described poet-programmer who loves building community through critical-creative code and bringing Intersectional approaches to AI. They are currently a Professor of Computational Thinking and Aesthetic Doing at Kunsthochschule für Medien Köln (Academy of Media Arts Cologne) and hold a PhD in Media Arts and Practice from USC.