Super-questioning

Consider a question such as ‘Will the UK have a trade surplus by the end of the year?’, how may you go about figuring out your answer?

Usually, you'll want to break down the question into multiple parts, then ask more questions related to each of those parts, each coming with its prediction. Prioritising the questions that matter the most to the main question first, and working your way down the tree.

For predicting the future accurately, you want to make sure that these questions are cross-domain. What you'll inevitably end up with, is an ever-increasing fractal of potential predictions that all link to one another.

We will generate 10^12 questions a month, synthetically made from real-world events and questions that humanity has placed its fate in.

The role of Super-questioners is therefore to take any future event, questioning what should be asked to know the answer, prioritising those they estimate would have the highest impact on the target event. Then creating a question set that is ever-evolving for predicting the outcome. It allows the Hyper-forecasters to branch down a logic tree of a prolific amount of questions that may only be tangentially related.

Sequentially going through and understanding what questions may relate to which others, leads to a more complete picture when answering questions on future events.

Moreover, they act as our 'event generators'. Much like in prediction markets such as Polymarket, we need events to be formally asked with correct sourcing. If an event is resolved with an unusual source, or the wording isn’t completely exacting, it's possible for questions to be answered differently due to a small conflict over the interpretation of a word or otherwise. This limits modern prediction markets to man-made and vetted event generation, leading to a tiny slice of the world.

Fortunately, the Infinite Game isn’t limited in the same way, as it can process questions with infinite granularity while being incredibly robust to outliers and randomness.

This gives us an abundance of questions and answers, an event landscape so large that we can start to answer events with long potential durations. It'll allow us to map out the 'human tech tree', that of learning what we should build and when, and how long it'll take us to achieve advancements in science and technology. Seeing far into the future on what we unlock and how we go about unlocking it.

These questions, however, need a predictor. Something or someone, to take them and go out into the world and try to answer them.

That's where Hyper-forecasters come in.