Matthew Adelstein makes a case for thinking about the welfare of digital minds:

Why think this? First, digital minds could be produced with amazing efficiency. Computing has grown vastly more efficient. If digital minds are possible, it should be cheap to create them in enormous numbers.

Second, in the far future, we should expect access to lots of energy. In the next many billions of years, we are likely to find ways of harnessing enormous amounts of energy. Nearly all the stuff in the universe that could be converted to value is outside of Earth.

It is a lot easier to convert space resources into computing than into biological organisms. Biological organisms need a stable habitat. Probably we’d need to terraform other planets to make them habitable for biological organisms. Yet this is costly and inefficient. For this reason, it would be hard to sustain very large populations of biological organisms. Not so for digital minds, which could be directly generated from space’s abundant resources.

These two factors mean there could be lots of future digital minds. One estimate put the number around 10^58, many orders of magnitude more than the number of humans, and there are reasons to think that estimate might be conservative. Don’t take this number too literally, but the important takeaway is that even if you were pretty sure that there couldn’t be digital minds, still roughly all expected future minds would be digital.

Earth is small compared to space (SOURCE???). In total it looks like we’ll be able to reach between 10^21 and 10^23 stars. Most welfare will be had, therefore, by the organisms that will populate space, which can by and large only be digital minds.