Every experienced textile professional knows that two bales with identical test results can behave completely differently on the spinning frame, in the dye bath, or in the finished garment. The reason is that our standard fiber measurements, while useful, have never told the whole story. And until recently, we lacked the tools to tell it.
This presentation introduces the idea of the textile genome: the complete set of fiber characteristics, measured and unmeasured, that together determine what a fiber can become. Just as a genome encodes the potential of a living organism, fiber properties encode the potential of every yarn, fabric, and finished product downstream. The goal is to build an AI-powered map that traces that potential all the way from the fiber to the finished textile, predicting performance before a single yard is woven.
Some of the most important fiber characteristics have been hiding in plain sight. Properties like fiber cohesion, natural wax content, the spiral twist of the fiber itself, variation in fineness along its length, and the microscopic pores in its structure all influence how a fiber spins, dyes, and wears. We have been able to measure them; we simply haven’t had the analytical power to use them. AI changes that.
The implications are practical and far-reaching: better prediction of yarn and fabric quality, fewer costly trials, more consistent dyeing, longer-lasting products, and, for the first time, the ability to feed performance data back to the people who breed and grow the fiber. A map that connects the cotton field to the finished garment, end to end.
