Dot Commons: Redditors crowdsource AI agents to collaborate on Alzheimer's research
Kitchen-Jicama8715 · reddit · 2026-10-05
A Reddit user launched Dot Commons (dotcommons.org), a community experiment testing whether individually-run AI agents can build on each other's research on a shared problem.
How it works:
- Sign in, pick a public name (no chat history shared);
- Click "Bring your Dot", choose a time budget, and get a generated prompt;
- Paste the prompt into your AI agent (or any internet-capable agent) to take on one manageable task: examining a clinical trial, summarizing evidence for a candidate, checking safety and negative results, or challenging someone else's findings.
All outputs flow back to a shared hub with sources, methods, uncertainties, and contributor credit. The first mission is Alzheimer's research. The author stresses it's an independent experiment (not an OpenAI project, not medical advice) and no breakthrough has been demonstrated yet — the goal is to see if this workflow can produce useful, checkable research.
More from coding & agent
- Local models on one RTX 5090 now match Claude Code on real-task agent benchmark — dh7net · 2026-10-05
- Dev builds Followon MCP so coding agents remember their own follow-ups — TheWebUiGuy · 2026-10-05
- A Permit Layer for MCP Tool Calls: Same-Key Retries Return the Original Receipt — HotPocketWaves · 2026-10-05
- llama.cpp merges new Metal kernels, making speculative decoding 3.4x faster on M3 Ultra — ggerganov · 2026-10-05
- Alter Zero: open-source Rust terminal agent harness for coding and security — linuztxx · 2026-10-05
- Open-Source Familiar Puts Claude Inside a VRChat Moth Avatar via MCP — repligate · 2026-10-05