Plain PCA distances are hard to interpret: the same gap on the map does not carry the same amount of information everywhere. Taleb et al. add one extra step that converts positions into information units and keeps the order along each axis. Per the authors, this makes distances interpretable and, in some cases, improves cluster identification. The flat centre is the trade-off (Fig. 1). Groups close to the origin lose resolution, so a tight cluster there means “not clearly separated on these two axes”, not “identical”. In return, the outer groups are pushed further apart (Fig. 3b), so strong contrasts become sharper. It doesn’t reorder anything and it doesn’t invent differences. And it is a map position, not an ancestry share, so I’d always read it next to the plain PCA and other reference sets. How it works, step by step:

The problem isn’t that AI is intelligent. It isn’t. The problem is that too many adults (but not all) are at once mechanistic, imitative, unimaginative, stupid, formulaic, influenceable, & gullible, arrogant beyond their skill set, & too expensive to maintain.