But what in the philosophy of EA really disallows it? In a utilitarian model, if I can do a lot of good by doing some bad (eg, steal from crypto investors and give the money to AI safety research), what did I do wrong? I appreciate that this can discredit EA and hurt their efforts, but this is a short-termist concern; in the long run, the AI research might end up being more important than the transient reputational damage. Furthermore, what if there's only a 5% chance of being caught doing something immoral? Then, in expectation, the positive impact is even higher.
It's all very well to pay lip service to 'common sense morality' but it doesn't seem to fit into the actual theology of EA, and this is what fanatics are most likely to follow.
It's all very well to pay lip service to 'common sense morality' but it doesn't seem to fit into the actual theology of EA, and this is what fanatics are most likely to follow.