01
Community conduct
Technical disagreement is expected. Personal hostility, exclusion, or status games are not.
Policies
The Semantic AI is an open collaborative initiative for semantic AI systems. These founding policies define how work is reviewed and governed as project volume and organizational surface area expand.
They cover collaboration norms, attribution, research integrity, deployment responsibility, and incubation governance.
02
Contribution and review
Work should be reviewable, reproducible where possible, and tied to explicit ownership.
03
Research integrity
Claims must distinguish measured results, informed inference, and future intent.
04
Attribution and IP
Code, notes, and benchmark contributions should keep authorship visible and licensing explicit.
05
Safety and deployment
Operational impact, abuse paths, and failure modes must be part of system design, not a postscript.
06
Incubation governance
Projects in incubation should have clear goals, maintainers, and exit criteria into independent surfaces.
Implementation Notes
What these policies mean in practice.
- Major technical claims should point to code, benchmarks, or written records.
- Incubated projects should state scope, maintainers, and what “ready” means.
- Operational tradeoffs should include latency, cost, failure handling, and human override paths.
- Shared outputs should preserve attribution across repos, notes, and downstream reuse.