That design wouldn't integrate with derailleurs too well. Maybe ok on a track bike as per the pic's.

Can you see Shimano or Campag licencing a design from someone else, for what is a core part of what they do ?

Maybe maybe IF it's more efficient then ok for track at the very very pointed end of racing. I look forward to the independent testing report being published !!

For e-bike ? They're clutching at straws there big style.
Who riding a fat heavy e-bike gives an F about losing probably half a watt ? If it's that important they'll have chipped the controller already and have 400 watts and no speed cut out rather than the regulatory 250w limit and 16mph cut.

for single sped use yup makes sense and for olympic track event absolutely where every watt counts. On derailleur systems??? the big three have no reason to go there, I doubt Shimano would ever license this, while if it was SRAM they would buy it outright would be the guess based upon past performances. I just think this is a fizzle for derailleur system in the long run.

If I look at the upper picture in the article my first thought is that you can refrain from the outer links in doubling the amount of inner links. You mount them not to the middle of the cradles, but to the end of the cradles aligned to the pins of the cradles. Every cradle is then connected to four links which have half of the length of the ones shown at the picture.
Because now the chain seems indeed too broad to be used with a sprocket-set.