“detect a shift of a trillionth of the diameter of a proton.”
You’ll be comforted to note that elsewhere in the article it says they only need to detect a whopping “few thousandths of the width of a proton,” on the order of a trillionth of a micron.
A micron is still sizable in many ways. A person can feel texture features down to about 1/100 of a micron in size.
My machines here have a repeatability of 1 micron. Unless the sun reflects off a shiny surface too long, or the idiots leave the door open on a cold or hot day, or a hundred other variables.
To detect a shift as small as they are referring to mind blowing to someone in my trade.
The official Caltech support site to my local LIGO (Hanford), says down to 1/10,000 the width of the proton charge nucleus. Since that web material gets reviewed by incoming graduate students all the time it is probably correct. The other statistic that amazes me is the vacuum inside the 4km tubes at a pressure lower than interstellar space (and maintained for long periods).