On the subject of the dangers of deepfakes, the most recent episode of The Blacklist addressed deepfakes in a storyline I found quite interesting. Basically (s

On the subject of the dangers of deepfakes, the most recent episode of The Blacklist addressed deepfakes in a storyline I found quite interesting. Basically (spoilers ahead), this researcher creates a sentient AI and the AI promptly decides that sentient AI is a danger to humanity and tries to kill a few of the top AI researchers. Ok, kinda unrealistic. The more realistic part? To get one of the AI researchers killed, a deepfake video is created of that researcher saying something along the lines of "over the years at X corp I've seen the worst of humanity, too much evil, it's time to end it all" accompanied by him strapping on a bomb vest. The video is released and everyone freaks out. He doesn't notice, goes to work, gets surrounded by cops while holding a small black device (his phone), and the police shoot and kill him thinking it's a detonator. I'd always considered deepfakes in the context of making false political statements which could eventually be disproved. Worst case, a bunch of people think the wrong thing for a while. This use case of forcing rapid response without time for validation or refutal is quite a bit scarier and one I personally hadn't considered before.