Are you not seeing the progress in the articles? It went from a lab proof-of-concept to a working prototype producing in real life 5000 liters/day passively. That's impressive as hell.
Colorado river averages 500,000 liters a second, and we use every drop of it. Scaling up from 0.055 liters per second is going to be expensive
Right. Remember, the issue is not whether it can be done, but how cheaply. Desalinization works fine now, but it's kind of expensive. These claimed breakthroughs are cost reductions. For that, you have to scale up to at least small production and measure costs. Only then you can boast.
2021-2023 is one approach, and 2024 is something else entirely. The 2024 thing is brackish groundwater cleanup.