IMO some GMOs are potentially useful: those which build up resistance to a pest and thus allow for LESS use of pesticides and thus less ecologic damage than with non-GMO varieties. Eg for cotton, a crop cursed with many many pests, GMOs can be great. Burkinabe farmers are (so far) quite happy that their government approved Monsanto's Gt variety some years ago, which I think is one of these.
Then there are other GMOs engineered to resist not a pest, but a particular herbicide produced by the same firm, eg glyphosate for Monsanto. The idea here is to be able to spread MORE herbicides into the environment, without hurting your crop... I totally reject these, as it is SCIENTIFICALLY well known that pesticides have many longterm effects on the environment (and perhaps also on sperm counts, and generally human health).
Other than that, i think some GMOs have been shown to induce allergies. Not too sure of the strength of the evidence there. There's also the risk of a transmission to a wild species of some resistance to an herbicide or another, but that is unlikely and wouldn't be a big problem.
I am not changing the subject. You are not the only one able to pick examples, or are you? And nuclear power also kills and pollutes. Try and think beyond the example, to what it means or illustrates.
My point is: don't confuse science and technology. Just because we can make some new thing based on scientific knowledge doesn't make it 'scientific', or necessarily good.