[arghwhat]

They describe it as a feature in their later 1942 patent (US2474994A), in the description of which they mention it to protect against powertools with poor torque control, claiming that a better screw would only be possible if power tools and their torque control were perfect and that it is therefore a required feature.

The original designs did not reference this though, so it seems more like them adopting an observed behavior as a feature - so the statement that it was originally designed to be that way is indeed off, but the company did adopt the behavior in official paperwork.