We are exhibiting this game about jailbreaking LLMs at the ACM Conference on Designing Interactive Systems 2024 in Copenhagen in a couple of weeks, and there's some really interesting research behind the game and why we designed it this way: 1. To heighten awareness of the risks of deploying fragile LLMs in everyday systems, 2. To heighten people's self-efficacy in interacting with LLMs, and 3. To discover the creative problem solving strategies, people deploy in this novel context. Pre-print here: https://lnkd.in/dzcvYVjF (All credit for the video teaser goes to Nanna, the video magician)
Your favorite co-author's favorite co-author | Cognitive Cost of AI • Human-Computer Interaction • LLM Security |
COMING. SOON. HACC-MAN - an arcade game for jailbreaking LLMs Nanna Inie, Jeanette Falk and Matheus Valentim Paper here: https://lnkd.in/dsWaZ2Wn
Super coolstuff, Jeanette! I look forward to reading the paper
Consultant | Expert in AI for Video Games
4moI love the teaser trailer! The paper and game also sound super interesting.