I suspect that the NCVT works by using the user as a reference, which under normal circumstances should act like a ground of sorts.
I think if you were isolated and touching a hot chassis with one hand and used an NCVT with the other it wouldn't light up.
If I'm correct, then putting an NCVT into any device would require a known good ground for it to be useful, otherwise it isn't any better than a neon bulb.
I could be wrong..
Jason
While I haven't designed one yet, and probably won't, I did think it through.
Indeed all voltage is relative. My hypothetical "smart" outlet strip would presumably have some products plugged into it. With none of those products energized, their mains cords should all be pretty quiet. Using a high impedance voltage detector to measure for voltage between outlet ground and all the mains plugs (H+N+G) we should be able to identify a hot outlet ground lead. About the only scenario that would create a false OK is if one or more of the unpowered mains cords had a hot voltage present too.
This scenario is extremely unlikely, but possible, I guess. For extra insurance I could poll between the multiple product mains cords plugged in separately. It seems even less likely that "all" mains leads would also have hot grounds too. So if significant voltage is detected between outlet ground and "any" one mains cord ground that outlet ground is disconnected, and the red LED starts flashing. If the outlet ground is hot and the product ground is ALREADY hot I didn't make the situation worse by connecting them, and the GFCI should still disconnect power if it detects leakage.
So I believe it could be done, but like many of my ideas, the world is not likely to beat a path to my door to buy one, well many just one to Jason to prove it works.
JR