> But there's consensus among the scientific community that it's being widely under-reported
Define consensus.
I used to work at Our World in Data, so I am biased, but I would consider them the most trustworthy curator of data sources (for one, they are open source and open to pull requests). So while there indeed may be problems with the India data (and we were constantly switching to better sources as they appeared), I would trust their data sources more than any one person on YouTube.
There are local regional(state) media trying to count the uncounted deaths like these -
BBC report on that: 'India's Covid crisis: The newsroom counting the uncounted deaths' - https://www.bbc.com/news/world-asia-india-56969086 . I wonder whether Our World In Data would accept these data, If so it would be worth a PR then again its just the data of a state.
Define consensus.
I used to work at Our World in Data, so I am biased, but I would consider them the most trustworthy curator of data sources (for one, they are open source and open to pull requests). So while there indeed may be problems with the India data (and we were constantly switching to better sources as they appeared), I would trust their data sources more than any one person on YouTube.