Where this is headed.
The AI frontier is advancing on raw capability while the reasoning underneath it stays unsound. Large language models pattern-match: they produce an answer one token at a time, and the chain of thought shown alongside it is a narration, not the computation that generated the answer. Research into chain-of-thought faithfulness has repeatedly found the two come apart, so the explanation a model gives is not a reliable account of how it actually reached its conclusion. Scaling has made this more capable without making it more sound. You can buy more horsepower, a larger model and more compute, but the thought process, the part that decides whether a hard judgment is actually right, is not something the current paradigm produces.
The work that matters sits upstream of the capability race, and we build it from the bottom up. The deepest problem is the thinking itself: getting a model to reason in a way that is structured, grounded, and checkable. That is what Redstone is. It imposes a process from outside the model and enforces it in code: the model cannot skip steps, its claims and assumptions are tracked to an answer, its reasoning is checked by fresh reviewers rather than by its author, and the human is brought in where their judgment changes the outcome. The trace is the actual structure of the work rather than the model's account of itself. A sound, stable thinking process is the floor everything else stands on, and almost no one is building it.
On that floor, the next problem comes into reach: making the system yours. A system that reasons soundly with your input can learn how you decide, what you prefer, and what your judgment looks like, and accumulate it over time. The deliberation happens inside your own real work, and your rulings stay attached to it: what was rejected, what was corrected, and why you decided as you did. We believe that is the data that trains a better partner: a model shaped by how smart people reason rather than by the conclusions the internet already holds. And in the work that matters most, there is no test that can say whether an answer is right, so your judgment stays the final check.
This is the order we believe it has to happen in, and the order we build in. Get the thinking right, make its improvement verifiable, then train a model on the process itself. Instance B exists to solve those problems, on a foundation solid enough to hold the ones above it.