This seems pretty reasonable and a fascinating suggestion nonetheless:
So for milleniums (about 10 of them) humans hung out on a forested, unsubmerged Beringia. That means people lived there for anywhere from 2/3 as long to just as long as it’s been since the last ice age, or about as long as it’s been since the rise of agriculture. That’s a long time to call it a “bridge”.
It’s not Starostin and it’s not Greenberg, so those red flags can be set down.
Not too sure on computational phylogenetics. My first instinct is to think that it sounds like using computers to calculate something more complicated than just numbers, i.e. calling statistics social scientific reality. If the method can be trusted, Beringia standstill is a good explanation for the relationship between Na-Dene and Yeniseian. More on what genetics can and can’t tell us about language another time.
The Language Families
Na-Dene – Includes languages in Central Alaska and the Yukon, and Apache and Navajo.
Yeniseian – Small family in central Siberia, severely endangered, with the Ket language of 200 speakers the last one