While explicit knowledge can be captured, tacit knowledge cannot be captured—or at least, not in a very practical way.
This is because tacit knowledge is internalized over time as the result of real-world interactions: as an example, it is common for great engineers to have a sense for what a beautiful system looks like, but also have trouble spelling it out.
This is why, when learning, you typically start from explicit and move to tacit: once you’ve absorbed and internalized explicit knowledge via standardized formats, you then move on to tacit knowledge which is best taught through mentoring and experience.
David Autor calls this Polanyi’s paradox, after Michael Polanyi’s line “we know more than we can tell,” and argues it has historically been the main barrier to automating work. It’s also a limit for AI: a person absorbs tacit knowledge by being immersed in it, but an LLM only has what someone feeds into it. It reasons from a thin, legible slice of the context and lands closer to the standard answer, which is one reason AI absorbs work in a predictable order.