AI Goal Misalignment
Jeremie discusses the potential pitfalls of text autocomplete systems, highlighting the unintended consequences of training AI with unclear goals. Edouard emphasizes the challenge of instilling true objectives in AI, illustrating this with an experiment involving a Mario character that fails to adapt when its goal is altered. The conversation reveals the complexities of ensuring AI systems understand and pursue the correct goals in various contexts.In this clip
From this podcast
The Joe Rogan Experience
Joe Rogan Experience #2156 - Jeremie & Edouard Harris
Related Questions
Do AI systems have hidden desires in the episode Jeremie Harris: Realistic Alignment and AI Policy and the clip The Challenge of AI Objectives?
Can AI have complex goals as discussed in the episode Jeremie Harris: Realistic Alignment and AI Policy and the clip The Inner Alignment Problem, as well as in the episode AI Researcher Reveals How to Set the Right Objectives | Knowledge Project Podcast and the clip The Objective Paradox?
Can AI motivations be shaped as discussed in the episode Jeremie Harris: Realistic Alignment and AI Policy and the clip The Challenge of AI Objectives?