AI-assisted discovery, clearly labeled
AI is used to discover candidate business models, identify common variants, group similar ideas, and convert unstructured research into a consistent schema. AI makes broad coverage possible; it does not make every generated claim true.
One idea must represent a distinct operating model
We do not count simple wording changes as new ideas. A variant should differ meaningfully in customer, delivery method, capital requirements, regulation, skills, or economics before it becomes its own entry.
What every entry contains
- Customer and core offer
- Directional startup budget and weekly time bands
- Work mode, experience, employment fit, and risk level
- Vehicle, licensing, and operating considerations
- Potential advantages, constraints, and first validation steps
- Last review date and confidence status
Current confidence status: directional
The launch library is designed for comparison and discovery. Costs, licensing, competition, and market demand must be checked against current primary sources in the user's location before action.
How the library should improve
Future research cycles will add source-level citations, geographic availability, conflicting-source records, confidence scoring, expired-data alerts, and category-specific review rules. Low-confidence records should be visible for browsing but should not dominate top matches.
Editorial independence
Inclusion and matching are not purchased. Future affiliate relationships or advertisements must be clearly labeled and must not change the match score.