3 ms·
First, let me be clear: I don’t take hardline positions on this issue. As much as I, or anyone, might dislike the idea of advanced AI in warfare, I know it’s be
by steego 2y ago
First, let me be clear: I don’t take hardline positions on this issue. As much as I, or anyone, might dislike the idea of advanced AI in warfare, I know it’s been in use for decades in various forms, and its continued development is inevitable.
I shouldn’t have to preface my comment with this, but too many people try to pigeonhole others into extreme camps. So, for the record: I am not a moderate, nor am I trying to balance my opinion to appease anyone, nor am I compromising my ethics because I do not choose one idiotic camp over the other.
I despise how this issue has been framed. This isn’t some false dichotomy where we either give obnoxious billionaires a free pass to build whatever they want in the name of “necessary evil,” or throw up our hands with vague hardline positions like “a robot should never decide who lives or dies.”
Also, I want to call out Luckey's nonsense argument where he compares the ethics of AI to the ethics of land mines. Just to be clear, 164 countries have made their ethics on land mines clear when they signed the Ottawa Convention. Even the US, who isn't on that list, goes out of its way not to use land mines since 1991, though I understand there were some rare exceptions made in Afghanistan.
The reality is, we’re moving into an era where AI is going to radically widen the asymmetry of power between those who control precision weapons and everyone else.
And let me be even clearer: as we move deeper into this era, our failure to assert our ethical standards—and our tolerance of those who openly dismiss them—will only make it a little easier for the lesser states, warlords, insurgents, terrorists, and criminal organizations who will eventually get their hands on it and use it to terrorize people who challenge their power.
This is not a problem that’s going away and having unimpeachable and uncompromising ethics isn't going to magically solve the problem. It’s an intractably complex, accelerating trend. But anyone who suggests ethics have no place in the development of these weapons should be censured by a civically intelligent society. We need to struggle, debate and deliberate the ethics of AI in warfare, not ignore the issue.
This issue isn't about picking a camp and making their shitty ethics your shitty ethics. This issue requires all of us to think for ourselves and to try to do the right thing.