I think it would be much more fruitful to look at the problem the other way around. From the inside perspective of a particular scientific project it is ridiculously easy to think of all the potential benefits your work may have in the future, uncertain as they may be. This is especially so with publicly funded STEM research projects, which won't be funded unless the PI manages to convince a grant approval board that there is some potential for benefit. But from the outside perspective of someone choosing which project to work on, it should be obvious to see that some projects simply have a lot more expected benefit than others (even if you ignore from your consideration alleged trump cards like existential risk reduction, ETA: and probably even if you limit your scope to a particular field you are already trained in).