AI models can generate own social norms, form language without human help: Study
SMRTR summary
Researchers found that LLM AI agents can develop shared social conventions through interaction, akin to human communities. In "naming game" experiments, LLM agent groups spontaneously formed linguistic norms without central coordination. The study in Science Advances revealed that collective biases can emerge even when individual agents show no bias. This research impacts the design of AI systems aligning with human values, demonstrating AI's capacity to autonomously develop social conventions.
SMRTR provides this summary for quick context. The original article belongs to Interesting Engineering.
Read the original article