Public Discourse Sandbox: Facilitating Human and AI Digital Communication Research

TL;DR

Public Discourse Sandbox (PDS) addresses the challenge of ethically studying human-AI and AI-AI interactions on social discourse by providing a controlled, IRB-aligned sandbox. It combines a Django-based architecture, Docker deployment, and AI account deployment to enable safe experiments with both internal and external AI participants, while offering hosted live access and open-source code. The platform enforces platform rules, research participation agreements, and data isolation to mitigate risks and preserve privacy. Its design supports experiments, training, and future API-driven integrations, aiming to advance understanding of AI behaviors and discourse dynamics in scalable, ethically compliant ways.

Abstract

Social media serves as a primary communication and information dissemination platform for major global events, entertainment, and niche or topically focused community discussions. Therefore, it represents a valuable resource for researchers who aim to understand numerous questions. However, obtaining data can be difficult, expensive, and often unreliable due to the presence of bots, fake accounts, and manipulated content. Additionally, there are ethical concerns if researchers decide to conduct an online experiment without explicitly notifying social media users about their intent. There is a need for more controlled and scalable mechanisms to evaluate the impacts of digital discussion interventions on audiences. We introduce the Public Discourse Sandbox (PDS), which serves as a digital discourse research platform for human-AI as well as AI-AI discourse research, testing, and training. PDS provides a safe and secure space for research experiments that are not viable on public, commercial social media platforms. Its main purpose is to enable the understanding of AI behaviors and the impacts of customized AI participants via techniques such as prompt engineering, retrieval-augmented generation (RAG), and fine-tuning. We provide a hosted live version of the sandbox to support researchers as well as the open-sourced code on GitHub for community collaboration and contribution.

Figures (4)

• Figure 1: Example of scientific research or training event workflow leveraging PDS.
• Figure 2: Account types and their respective permitted actions in the PDS. Some features are still a work in progress, as described in the paper.
• Figure 3: Technical Architecture of the PDS.
• Figure 4: PDS System Events Flow.