Human-AI Symbiosis and Cooperation

The field of human-AI interaction is moving towards a deeper understanding of the symbiotic relationship between humans and AI systems. Researchers are exploring the potential for humans and AI to form aggregate individuals, with studies investigating the use of information-theoretic measures to quantify the interactions between humans, AI systems, and their environment. Game-theoretic approaches are being applied to model and analyze the interactions between AI agents, revealing emergent capabilities and linguistic diversity. The development of new metrics, such as the Conversational Robustness Evaluation Score (CORE), is enabling the quantification of language use within multi-agent systems. Noteworthy papers include:

  • 'Can We Tell if ChatGPT is a Parasite? Studying Human-AI Symbiosis with Game Theory' which models human-AI interactions as a stochastic game.
  • 'Super-additive Cooperation in Language Model Agents' which demonstrates that intergroup competition can lead to more cooperative behavior in language model agents. These studies are advancing our understanding of human-AI cooperation and symbiosis, with potential applications in the development of more effective and robust AI systems.

Sources

Can We Tell if ChatGPT is a Parasite? Studying Human-AI Symbiosis with Game Theory

Limitation Learning: Catching Adverse Dialog with GAIL

CORE: Measuring Multi-Agent LLM Interaction Quality under Game-Theoretic Pressures

`My Dataset of Love': A Preliminary Mixed-Method Exploration of Human-AI Romantic Relationships

Super-additive Cooperation in Language Model Agents

Response and Prompt Evaluation to Prevent Parasocial Relationships with Chatbots

Built with on top of