If extremely underpowered, i'd go 24/0. If it can get 'close enough' to 35-40 DLI before that point it becomes a debate... at what point does the negative of zero or just fewer dark hours have a larger impact than the incremental DLI you add? I don't believe solid information exists on such an esoteric question. The context is only relevant to people who didn't properly size their equipment for the space. So, not a high priority to test something that can be easily avoided.
"dark" cycle processes do happen in the light, too. You don't have to worry about killing the plant or anything like that. It's just not optimal to do it.
--------- sorta related but unnecessary to read if you are of the TL;DR generations since 1981.
I want to clear up soemthing mistated too. Stomata close at night, therefore transpiration is significantly slower at night. CO2 enters through the stomata. CO2 intake is not a major concern for the plant at night.
Some businesses may run lights 24/0, but they may also blindly be wasting a ton of electicty for no reason. If they were a professional operation with the right expertise, they wouldn't have underpowered lights that require a 24/0 cycle in the first place.
Whether you give X DLI over 24h or X DLI over 12-20h doesn't matter much as far as cost, either. It'll cost roughly the same wattage. You may save a fraction of a percent of watts by running already efficeint lights with a bit less wattage over longer hours, but you don't even need to do a break-even analysis to see it would not be worth it if it has any negative impact on the plants. So, if the plant benefits froma dark cycle, which it does, it's better to give it some minimum dark cycle and meet its DLI maximum over 18 hours of light operation per day. This stuff is all math and not magic.
I run my vege 16/8. also creates a less drastic shift when i do switch to 12/12, because i don't grow circus clown plants (autoflowers lol, i can't resist).