 Sora uses a version of the diffusion model used by Dolly and GPT for generates a video by starting off with one that is more or less just a bunch of static transforms it to make a finished product, which is surprising and delighting many people out there and also terrifying others. Yeah. I feel like these are, are very clearly going to have pretty close to the same evaluation as we've had of all these other tools like chat GPT, which is very impressive when you first see them. Probably not as scary as some people are worried they will be. And then useful for, for specific things. I was laughing because the truck looks good and the background looks good, but the dirt coming off the tires is completely wrong. Having spent hundreds of hours following, you know, desert vehicles, spraying dust like that. We've got the very warranted fear from people saying, oh gosh, this is my job. What's, you know, what's my job going to look like in another year? You know, this is all happening so fast.