Read before you write
Most of the excitement around brain interfaces is about writing: changing a state, lifting a mood, sharpening focus. We understand the pull. We think it is the wrong place to begin.
A system that changes how a person thinks has to first show, in numbers, that it knows how the person thinks. Otherwise the change is a guess, and the person cannot tell the guess from the effect. Reading first is not caution for its own sake. It is the only way to build the measurement that makes writing honest later.
So our first product only reads. It is deliberately simple. It gives a small amount of true context to a model that already knows how to use context. That is enough to be useful, and it produces the data everything else rests on.
When writing does come, it will not be a stimulus. It will be a point in a chain of thought where a sentence sets something off: a question, an analogy, a contrast, placed where the person has agreed to have it placed. Every such action carries its own effect measurement, and there are always windows without intervention to compare against.