Are we at a point now where Asimov’s laws are programmed into these fellas somewhere?

Nope.

The article clearly spells out that it's end to end LLM. Text and video in, motor function out.

Technically, the text model probably has a few copies, but they are nothing more than Asimov's narrative. Laws don't (and can't) exist in a model