Achieving Believability
Joon discusses the misalignment between generative agents and the objective of achieving believability. He explores the implications of using models that prioritize usefulness over conflicts and fights, and raises important questions about the potential benefits and dangers of creating believable simulations.In this clip
From this podcast
The Gradient
Joon Park: Generative Agents and Human-Computer Interaction
Related Questions
What are some non-obvious insights about the impact of multi-agent systems as discussed in the episode Joon Park: Generative Agents and Human-Computer Interaction and the clip Achieving Believability from the episode Inside AI Town: What AI Can Teach Us About Being Human?
What are some non-obvious insights about the impact of multi-agent systems as discussed in the episode Joon Park: Generative Agents and Human-Computer Interaction and the clip Achieving Believability?