Sam Altman's OpenAI Sora creates one-minute text-to-videos, but given deepfakes, how safe is it?

Sam Altman's new AI tool's videos are so convincing that it can fool you. It has realistic-looking two golden retrievers podcasting on top of a mountain, half duck half dragon, drones on Mars, animals riding bicycles. But the real creepy part is the possible fakes. These seemingly mundane visuals - like the cat walking through a garden or a dog climbing the windows - look so real that no one will really question their authenticity. So, in a world of deepfakes and misinformation, how safe is it?