Mercedes did in the Vision EQXX:
https://www.autocar.co.uk/car-news/new-cars/mercedes-benz-vi...
https://www.cnet.com/roadshow/news/mercedes-vision-eqxx-ev-c...
I guess 10kWh/100km is possible at 90km/h on cruisecontroll behind a big truck.
EVs biggest problem are the inefficiencies in the electric grid, there is like a 50% margin for improvement.
Also one could fit cars with rail gear and run them on rail tracks on the highway which would lessen the tire friction roll resistance losses. Technology for allowing cars to run on rail tracks already exists for railway maintenance cars. Self driving on rail track is also much easier to implement.
https://en.m.wikipedia.org/wiki/Road–rail_vehicle
Safety from car crashes would also be far safer at 80km/h than higher speeds.
Germany has a short section of highway testing that setup for trucks
https://www.sustainabletruckvan.com/pantographs-for-trucks-s...
If your 90% efficient drivetrain-tyre system uses 20 kWh per 100 km, then improving that to 95% efficient maybe reduces that to 17 kWh.
Pencils and javelins have lower drag coefficients than cars. Unfortunately people don't fit inside them. A javelin-shaped vehicle big enough to fit a person comfortably inside is a train.
or a cybertruck.
And if people cared about efficiency enough to drive at 80 they'd do it now. Low efficiency stsrt/stop driving is the problem to solve.
Probably not with normal-sized cars.
For the Model 3, the theoretical minimum is 13.8 kW. This increases linearly with drag coefficient and area (as seen from the front). You would have to build a significantly smaller (flatter, narrower) car and improve the drag coefficient (0.2 has been achieved) to get there.
Here's a calculator in German (sorry) for experiments: http://spare-benzin.de/Fahrphysik/Luftwiderstand/Luftwiderst...
https://lightyear.one/lightyear-0
Initial model is > $300k but they plan a $30k model in the future.