AI Agent for Good: Alignment, Safety, & Impact

2025 Summer Symposium Hosted by Stanford Deliberative Democracy Lab
Invite-Only

As generative AI rapidly transforms our world—much like the printing press, electricity, and the internet before it—its potential to empower individuals with intelligent personal assistants is matched only by the magnitude of its risks. The ability of AI agents to generate new content and take autonomous actions introduces challenges such as misinformation, polarization, and bias, raising questions about industry, government, and public readiness. Some see these dangers as immediate and serious threats to society, while others argue that such worries may be exaggerated or premature. This symposium is the beginning of a series of efforts led by the Stanford Deliberative Democracy Lab to engage the global public and industry leaders to foster thoughtful deliberation and collaboration on how to develop AI agents. 

The outcome of this effort will include actionable guidelines and best practices for building safe, impactful AI agents that benefit both businesses and society, culminating in open reports and industry-wide recommendations.

Learn more about other technology and AI governance Deliberative Polls here:
Meta Community Forums: 2022, 2023, 2024
National Deliberative Poll in Taiwan on Information Integrity 
AI in Schools
Internet Governance Forum

Invited Call for Papers

Submission Window: April 21 – June 13, 2025

We invite industry leaders, researchers, technologists, policymakers, and practitioners to submit short papers (1–3 pages) for the upcoming AI Agent for Good: Alignment, Safety, and Impact symposium hosted by the Deliberative Democracy Lab at Stanford University.

Generative AI and autonomous agents are rapidly transforming the way we work, communicate, and make decisions. While these technologies promise unprecedented opportunities—intelligent automation, personalized services, and new creative frontiers—they also present urgent challenges around safety, alignment, transparency, and ethical use.

This symposium and subsequent deliberative forums will gather voices from across sectors to explore how we can develop AI agents that are safe, responsible, and beneficial for society.

Topics of Interest

We welcome submissions that address (but are not limited to) the following themes:

  • Frameworks for aligning AI agent behavior with human values
  • Risk mitigation strategies and safety guidelines
  • Use cases demonstrating responsible AI agent applications in business or public domains
  • Best practices for human-in-the-loop design and governance
  • Technical, ethical, or legal challenges in AI agent deployment
  • Cultural or regional perspectives on the social impact of AI agents

Submission Guidelines

  • Length: 1–3 pages (PDF format)
  • Deadline: June 13, 2025
  • Submission opens: April 21, 2025
  • Acceptance notifications: Rolling basis (no later than June 21, 2025)
  • Symposium date: July 30, 2025 (Wednesday)
  • Submit to (and send any queries to): deliberation@stanford.edu

Authors of selected papers will be invited to participate in the in-person symposium at Stanford University in Summer 2025, and their work may be included in published proceedings and cross-industry guidelines on safe AI agent development. The symposium is intended to be a small and focused gathering. 

We encourage contributions from startups, enterprises, academia, government, and nonprofit sectors. If your work involves advancing the safety, alignment, and real-world impact of AI agents, we want to hear from you.

 

 

Photo Credit: Growtika on Unsplash