> It's not possible to predict the uses of one's work Probably not in all cases, but you can often make an educated guess. You could at least consider classes

> It's not possible to predict the uses of one's work Probably not in all cases, but you can often make an educated guess. You could at least consider classes of activity, ranging from: Primary, sole purpose is to cause direct harm. You'd have to be dumb or willfully ignorant not to realise. "This BabyFaceMelter technology will be so amazingly terrible nobody will ever want to fight us again!" High likelihood of indirect harm - nominally for other purposes, but trivial to see simple ways in which it can be weaponised/made harmful. "Our Public Order Droid platforms are all equipped only with non-lethal electro-tickle cannons, and it'd be really hard to put actual bullets in there" Moderate likelihood of indirect harm. Fairly compelling "good" uses, but still possible to think of ways it could go bad. Less toxic or higher power explosives intended for mining, perhaps. Too generic to really know - finally, the weakest level that you present, something like better off-road vehicle tyres. Quite a lot of fundamental science (ie: not the "we hypothesise that X will make the Anthrax really angry" sort) has such wide application that you can't really guess. Ditto basic algorithms in maths & CS, or software libraries (is OpenSSL evil?) I don't totally disagree with you, but you might want to try painting with a slightly smaller brush to make your argument more compelling.