TED Talksのインスタグラム(ted) - 8月18日 00時14分

Think you could spot a fake video? Don’t be too sure. Computer scientist Supasorn Suwajanakorn developed a way to combine AI and 3D modeling to create shockingly realistic fake videos of people synced to audio. Using only existing photos and videos, he can build a moving 3D model of a person’s face and even replicate their speech. While there are some creative opportunities with an invention like this (for example, recreating lectures from some of history's most inspiring teachers or enabling authors to read their books aloud), there are also many grave ethical implications. That’s why Supasorn is working on tools to help prevent misuse. “If we use it right and carefully, this [technology] can allow any individual's positive impact on the world to be massively scaled and really help shape our future the way we want it to be,” he says. To learn more, watch his #TEDTalk at go.ted.com/fakevideos

[BIHAKUEN]UVシールド(UVShield)

>> 飲む日焼け止め！「UVシールド」を購入する