A simulated AI conference featuring diverse agentic personas delivering perspectives on "AI in 2026: Looking Forward"
This project explores perspective synthesis through agentic AI—using AI agents to simulate a conference where diverse personas (HR professionals, CEOs, doctors, artists, skeptics, enthusiasts, and many more) each deliver a 3-minute speech on AI's impact from their unique vantage point.
Unlike a panel discussion (which requires complex rebuttal orchestration) or a think tank (which requires institutional memory), this is a clean, one-time event: an orchestration agent distributes prompts to sub-agents, collects their speeches, and compiles the output.
🚧 Design Phase — Defining agents, methodology, and implementation approach
- PDF: Concatenated speeches with agent introductions
- Audio: TTS-rendered conference with MC introductions and voice diversity
├── design/ # Design specifications and agent definitions
├── notes/ # Original voice notes and transcripts
├── output/ # Generated conference outputs (future)
└── src/ # Implementation code (future)
This project is part of a broader exploration of agentic AI for perspective synthesis:
- AI Panel — Multi-agent debate with rebuttal mechanics
- AI Think Tank — Persistent institution with memory and publication workflows
- Agentic UN — Simulated General Assembly with structured voting
- Peace in the Middle East — Geopolitical ideation challenge
Open source — method and results shared publicly.
For more Claude Code projects, visit my Claude Code Repos Index.