Trust ledger
The engine should be judged too.
IdeaForge is designed to be challenged. Scores include reasoning, confidence, and follow-up surfaces so the product can learn where its judgment was sharp and where it was wrong.
Public predictions
Opt-in analyses can publish falsifiable predictions so engine confidence can be compared with crowd judgment.
Idea graveyard
Outcome follow-ups at 30, 60, and 90 days create a record of what actually happened after the report.
Anti-portfolio
A page for misses, false negatives, and lessons from ideas the engine judged incorrectly.
Leaderboard
Opt-in calibration rankings let founders compare judgment quality, not just idea quality.