KeepRI: Independent Reasoning in the Age of AI

Field notes

KeepRI describes a two-track model: free reasoning practice alongside optional cash-prize entry tied to a research agreement. Its proposed data operation would collect learning records, game environments, and human evaluations for AI teams, with separate consent and commercial licensing. The homepage names six principles—including independent judgment, learning through challenge, and participant choice—and says research enrollment is not active while cash-prize events remain in development.