Or it reeks of "turns out, building cars at scale is really hard, and we'd have to compete directly with experienced companies for uncertain gain". Partnerships or M&A are a much better way to go here then software companies trying to build car factories from scratch.
But making, marketing and distributing 10's of thousands of cars is so outside the scope of anything that Google is good at ...
Cars are crazy regulated, massive supply chains, complex distribution networks, massive finance issues both in terms of production but also consumer financing, massive labour issues, massive and long term capex in plants and equipment ... it's just not Googley.
I suggest that it's probably the right thing for them to do.