Transcript detail
Loading...
Public transcript context with linked callsigns, related nets, and analysis metadata.
Transcript
Public transcript text
in a blue jeans and a white jacket standing right next to my bike, that very first scene is a real image. It's a still frame from a video. So I took that still frame image, sent it to AI and said make a 12 second video of me magically appearing in a racing outfit getting on the bike and riding away. So that's what it did. And it's quite accurate. If you look at that carefully, if you look in the mirrors on the handlebar, you will see that the flame around me is reflected correctly in the mirrors. So it did a good job. But when it pulls back, it estimates what my bike looks like because it never got a full shot of the bike. So when I'm riding away, that's not what my bike looks like. It just did an estimate of it. So it, but it still looks good in my opinion.
Explore