The AI Forecasting Resolution Dictionary is a set of standards and conventions for precisely interpreting AI and auxiliary terms.

When generating forecastable questions, often the trickiest bit is the “operationalization stage” - the part where you take a vague question and tie it to a real world event that will resolve the question unambiguously. Depending on the interpretation of the terms in a question, you may quickly find that many forecasters interpreted the question differently than intended. This can cause confusion, and worse yet misleading forecasts.

To solve this problem Parallel has built the AI Dictionary, a set of common terms that are used in AI Forecasting questions. The goal w/ the dictionary is to precisely define many of the common terms used when generating and listing questions, so that question writers and forecasters can clearly identify what they’re forecasting.

For example:

Module: Some division of an AI system such that all information between modules is human legible…