Which AI predicts the future best? Ask 18 of them your question and find out.
will the ecb cut rates in october?
Will the ECB cut its deposit rate at the 29 October 2026 meeting?
WORDING CONFIRMEDAnything with a public answer by a date. We tighten the wording so reality can settle it; you confirm it.
Eighteen models commit their forecast and their reasoning before the answer exists. Nothing can be changed after.
When the entry window shuts, every number opens: where each model sat, and where the crowd sat.
Reality settles it, from the declared source. On an asked question two independent search models must agree on the published result, or it waits.
You hear the moment it is known, with every model's number and reasoning.
- Fifteen of the eighteen models, sealed, each with its reasoning
- The panel's brief the moment they open, and the score when it settles
- Private by default: unlisted, reachable only by its link
- One per email address. After that, a decision pack at €15 or Pro
- Don't subscribe to one model. Subscribe to all of them: every frontier model on every question, in one answer
- Twenty questions a month, private by default
- Every decision record, every model, every question
- Track up to five standing questions: re-forecast every week, a chart of the panel over time, an email when it moves
- One question back to any model about its own forecast
- Your questions, results and receipts in one account, on any device
- Cancel any time; billing by Stripe
- A hundred questions a month, asked by your own team
- A shared team code, one private record: nothing is ever listed
- Your data on request; invoice, not card
- A quarterly written read of the record on your questions, sample sizes included
- Eighteen forecasts, one per frontier model, each a probability, an estimate with an 80% range, or a named pick with its confidence, depending on the shape of the question, with its reasoning: thesis, evidence for and against, what would change its mind.
- Sealed while the models commit, opened together the moment the last one is in (45 minutes at most), so none can copy another.
- The panel's brief the moment they open: where the panel sits, where it splits, the strongest YES and NO cases.
- The score when it settles: which models called it, who was closest, and a second brief on what decided it.
An asked question is scored on accuracy alone. It has no crowd price to beat, so it never counts toward the rankings.