Here is an excellent essay by Peter McLaughlinhere is an excerpt:
So the problem is this. Effective altruism means being able to say that things other than utility matter – not just in the sense that they carry some moral weight, but in the sense that they may actually be relevant to deciding what to do, not just overwhelmed by utility calculations. Cowen clarifies the condition, identifying it as the denial of the following assertion: Given two options, no matter how other morally relevant factors are distributed among the options, you can still find a utility distribution such that the option with the highest amount of utility is the best. The hope that you can have “utilitarianism minus the controversial” rests precisely on denying this claim.
This condition is not intended to render the usefulness off topicso that utilitarian considerations should never make you change your mind or perspective: it is enough that they can be restrictedhelpfully coexist with other valuable purposes. It ensures that usefulness will not automatically overwhelm other factors, such as partiality to family and friends, or personal values, or self-interest, or respect for rights, or even suffering (as in the very disgusting conclusion). This would allow us to respect our intuitions when they conflict with utility, which is precisely what means to be able to get off the train for a crazy city.
Now, at the same time, Effective Altruists too want to emphasize the relevance of the scale for moral decision-making. The central idea of the first effective altruists was to resist range insensitivity and begin to systematically examine the numbers involved in various issues. Effective “long-term” altruists are deeply motivated by the idea that “the future is vast”: the large number of future people who could potentially exist gives us many reasons to try to improve the future. The fact that some interventions produce so much more utility—do so much more good – that others is one of the main reasons for prioritizing them. So although it is technically a solution to our problem to state (for example) that utility considerations effectively become irrelevant once the numbers get too large would be unacceptable to efficient altruists. Ladder Questions in Effective Altruism (rightly so, I would say!), and it does not cease to matter after a certain point.
There is much more to the argument, recommended.