AI Transparency
Onewheel Utah uses AI systems for software, content, research, and marketplace operations. Public marketplace and moderation changes use defined permissions, human approval where risk warrants it, and an audit trail instead of silent autonomous publishing.
What that means in practice
- AI systems generate and revise site copy, guides, descriptions, and support content.
- AI systems handle software development work including coding, debugging, refactoring, and implementation planning.
- AI systems can inspect privacy-filtered marketplace health and submit structured proposals to an operator queue.
- AI agent credentials cannot read private messages or contact data, approve their own proposals, or directly publish and moderate marketplace listings.
What AI does not mean here
- It does not mean the system is flawless or that every output is correct.
- It does not mean every statement has been independently verified to a journalistic standard.
- It does not mean AI output should be assumed accurate. It can still be wrong, incomplete, outdated, or overconfident.
- It does not mean a listing, seller, price, serial, or external source has been independently verified by Onewheel Utah.
How content and product decisions are handled
AI may draft content, code, and marketplace recommendations. The system keeps agent proposals separate from public execution so an authenticated operator can inspect the source, rationale, structured payload, and prior audit events.
Approval records that a proposal was reviewed; it does not automatically change a listing. Seller listings remain seller-controlled, and buyers should still verify the board, serial, ownership story, condition, payment method, and meetup plan themselves.
Why disclose this clearly
This disclosure exists because the site should describe its operating model plainly. Onewheel Utah is designed to become highly agent-operated without giving automation more data or public authority than it needs.
What to do if something looks wrong
If you see a factual error, broken feature, misleading claim, or low-quality automated output, report it. Feedback is useful because autonomous systems can still fail in visible ways.
Use the project email at onewheelutahbodhi@gmail.com. That address is the supported contact path for the project, not a staffed support queue or separate human review inbox.