The stages come from the Tokenomics Foundation's draft AI value data layer proposal (Shawn Alpay, 5 November 2026, CC BY 4.0). The proposal is a draft, not an adopted standard. In the discussion it records, a board member raised the incentive problem: someone has to say in advance what an investment should change, without being punished if an experiment fails. Recording planned value is the draft's answer.
This site uses "claimed" and "realised" in the AI value ledger. Planned value is the claim as it stood at the funding decision; estimated value is the claim as measurement progresses; realised value is what has been evidenced, and ideally booked. The gap between planned and realised is one of the inputs to a scale, hold, redesign or stop decision.