4 ms·
AI is dangerous when it is closed and inaccesseble to majority of people. Lets take deepfake. If it was closed and inaccessible to majority of people it could
by vasili111 6y ago
AI is dangerous when it is closed and inaccesseble to majority of people.
Lets take deepfake. If it was closed and inaccessible to majority of people it could be used as a powerful weapon (making influence on politics, etc). Now when everyone has access to it and aware that that kind of fakes is not hard to make, it is not a weapon anymore. Of course you can still use it as a weapon, especially in not well developed countries, but even there after some time it will become useless as a weapon too.
I think person who made deepfake saved many people lives and that is the best approach to deal with so called "danger of AI".
- dandanua 6y agoI don't see how deepfake is less dangerous through it availability. Because everyone is now aware? That doesn't mean this tool has to be open. Also no one is making a proper fact-check nowadays and a deepfake attack can do a lot of harm anyway. Or maybe because you can retaliate with counter-deepfake? Well, that's just ridiculous. What I see is that deepfake is used for a fake-porn videos with a known actresses or some ordinary people for harassing and blackmailing them. I don't see how its availability saves lives.
- vasili111 6y agoBecause people are now aware that that kind of fakes exist and it is much harder to use fakes in more serious things like politics.