It makes a difference in aviation, but the petrol -> mechanical energy conversion of an ICE is so lossy that it makes barely any difference in a car.

My car's fuel efficiency barely changes when I go from just me in the car to 5 people in the car, but a EV's range will change substantially (I presume).

Why would you presume that? if a gasoline car doesn't get appreciable difference when there's the extra weight of 4 additional people in it, why would an EV, which is the same four wheels and aerodynamics, just with a different motor in it going to be any different?

Because the battery -> mechanical energy process is basically lossless while the petrol -> mechanical energy process is lossy in a way that is largely insensitive to the power output.

ie. an EV that needs to produce 1kW might use 1.1kW from the battery and when it needs to produce 2kW it might draw 2.2kW (~2x) from the battery. So 10% increase in weight might result in a 10% reduction in range.

An ICE that needs 1kW mechanical energy will probably burn 10kW worth of petrol; when it needs 2kW of mechanical energy it’ll burn 11kW (<2x) of petrol. So a 10% increase in weight would result in a <10% reduction in range.

It's true an ICE has non-linear output power, and a "sweet spot", but at freeway speeds the force required to move the car is dominated by air resistance rather than weight, especially considering how heavy cars are to begin with. Adding 10% mass only costs ~3.5% at freeway speeds.

Directionally, your numbers are right, but we'd be looking at more like 5 kW to idle an ICE regardless, and then 6 kW for 1 kW output and 7kW for 2kW output (efficiency going from 17% to 29%). Instead of doubling output though, we'd only need to go from 1 kW to ~1.035 kW if we increased mass by 10%.

Which brings us to the topic of the batteries being heavy AF, but that's a whole other question.