Sorry, I edited my earlier outcome for clarity. Here is the context again: > Lately one of my luddite friends keeps mentioning how computers will soon kill us

Sorry, I edited my earlier outcome for clarity. Here is the context again: > Lately one of my luddite friends keeps mentioning how computers will soon kill us all. Perhaps to add to his torment, I like to retort that computers won't need to kill us, they'll just need to convince us to kill ourselves. > Certainly there are many ways to affect that outcome via social media. I wonder what other tools the deepfake future will provide? Stealing the voice of a dead loved one could be used to great affect to push a troubled soul over the edge. Anyway, are we talking about changing the death from being caused explicitly by computers to users being motivated by computers to kill themselves? I think the comment is instead talking about there being myriad ways computers could manipulate their users into killing themselves. It depends on the predicate of "that", but one construction seems more consistent. Shouldn't the second "affect" also be "effect"?