DeepMind researchers find LLMs can serve as effective mediators
A team of AI researchers with Google’s DeepMind London group has found that certain large language models (LLMs) can serve as effective mediators between groups of people with differing viewpoints regarding a given topic. The work is published in the journal Science.