The most frequent question is "Can we really win?". The answer is "There is no AI that keeps winning."

Since launch, the most frequent question has always been this: “Can you really win?” At the end of this serialization, I’ll answer directly using what I’ve written here.

First, for any product, the answer “you will win if you buy it” does not hold. As written in earlier articles, there is no AI that keeps winning with current technology; when the market changes, any logic can break. If you ever see someone claim “you will surely win,” please discount that claim as a reason to retreat.

That said, this is as far as I can answer. In our tests, among 1,024 AIs, the survivors showed 227 outcomes, a win rate of 75.3%, and a post-deduction +530.8 pips (these are test-figure numbers and do not guarantee future results). These numbers were produced using recently updated data not used for training, via full automation with all outcomes as the denominator. And this AI will someday deteriorate; when deterioration occurs, it will be displayed on the performance panel, and you can retrain and reselect from scratch.

In other words, the precise answer to the question is this.We do not sell a AI that continuously wins. We sell the process of selecting the AI that currently works and reselecting when it stops working.Whether you win depends on the market and management, but at least we can guarantee a design that cannot be disguised with numbers and a means for you to rebuild on your own when things collapse. My honesty about the “truly” part ends here. And in this industry, not many products answer this candidly.