10% I believe was originally an arbitrary rounded number that shouldn't have too bad an impact on someone. It's not novel, by any means, as you can see Islam advocating similar with Zakat! The novel part of EA was analyzing the outcome of charities based on research to ensure it actually improves lives in a quantifiable way. A great example of why this is useful is [0], which tl;dr was a highly funded waterpump dressed up like a childs toy to replace manual pumps in Africa where kids could play and water would be brought up to the surface. The issues were numerous as explained in that post! Now look at an EA-aligned org called Givewell [1] and what sort of research they do to find the most 'effective' charities where that is lives saved (or formally, Quality Adjusted Life Years) per dollar assuming all lives are equal (if $1000 saves a life in Ethiopia compared to $10000 in the US, saving 10 lives is more 'effective' roughly speaking). That's the core gist of EA. It's not re-skinned utilitarianism, it's not some crazy plot to get rich, it's analyzing charitable giving like you would an investment. There are then groups like Giving What We Can that advocate one should donate a percentage of their salary to these charities and the outcome is you might save more lives working as a developer in the Bay Area than a nurse, over your 40 year career (instead of being a nurse yourself you could fund 10 nurses in Nigeria, for example).
[0] https://forum.effectivealtruism.org/posts/P9XEtnHFQF24HCC8v/...
[1] https://www.givewell.org/
