i also saw the same eurobsdcon youtube video. i agree it is somewhat hard to imagine replacing long-time unix tools with newer versions that only differ in being implemented in a new language. making sure nothing breaks takes so much time, with not enough reward.

but writing new tools in safer languages makes more sense to me.

yes, you would need the toolchain in base. not easy, but should be possible. i only know go. perhaps it's still too much of a moving target for openbsd.

you could write programs in new safe language and put them in openbsd ports. problem is that it then isn't a "part of openbsd". do openbsd developers really want to write their next daemon in c? for how long will they stick to c?