> Future AI Risk defense could become an even broader industry than anti-drone systems. This is cute, but from a pragmatic perspective I don't think any engine

> Future AI Risk defense could become an even broader industry than anti-drone systems. This is cute, but from a pragmatic perspective I don't think any engineers are expecting that. It feels more and more like today's AI pundits have their head so far up their ass that they could give themselves a tongue-powered colonoscopy. You can't secure yourself from drones. That's the struggle - if you are in any open-air space, you can be surveilled or even targeted by a loitering machine. That is a problem, one that's hard to enforce legally for adversaries, and one which requires active defenses to defend against drones when they're detected. AI isn't the same way. You can protect yourself from AI, both through legal incentive and by rejecting non-authorized users. Traditional cryptosystems are still well-defended against AI, the "cracks" come from social engineering that is technically available to crafty humans as well. Stopping the social engineering won't entail AI warfare, but the general application of security principles that have been around since Suzie Thunder.