Video generation is merely the starting point and prerequisite for what companies like Meta ultimately want: a 'dopamine generator' that wires human engagement signals (likes, views) directly into models to produce content optimized to keep people coming back.
Discussing Meta's new Muse Spark model, Pash argues video generation is just a prerequisite step toward Meta building a 'dopamine generator' that trains directly on likes and viewing behavior to maximize engagement. ✦ AI generated
Pash · The Cognitive Revolution · 2026-07-08 · original ↗
starts at this moment · 88:45
the video gen is just the starting point for the dopamine gen, right? Right? Like the the video gen is kind of a prerequisite to the dopamine uh generator. Uh but the dopamine generator is coming, right? And and that's the ultimate goal. Ultimate goal is not a video gen. The ultimate goal is dopamine gen, right?
verbatim transcript · starts at 88:45
88:45spit out things that, you know, cause dopamine and cause people to keep coming back to Facebook. So, and and that is maybe a signal that no other firm has as much as Facebook. I mean, other I I think Facebook and Tik Tok, right? These are two um and I think the video model, the video gen is just the starting point for the dopamine gen, right? Right? Like the
89:09the video gen is kind of a prerequisite to the dopamine uh generator. Uh but the dopamine generator is coming, right? And and that's the ultimate goal. Ultimate goal is not a video gen. The ultimate goal is dopamine gen, right? >> Yeah. Uh mixed feelings on that to put it mildly. But um >> every everyone everyone is scared of meta, right? like when when Meta decides to go for it because I think I think
89:36people in the industry kind of know what is possible but what most people most researchers have not focused on and you know having researchers focus on that or having the best researchers focus on that is something that is scary I think so but you know people have been talking about it for a while right it's not it's It's not it's not exactly, you know, a new idea.
90:06>> Yeah. Well, we've seen a version of it, right? I mean the the the sort of 40 sycopants, the the sycopant aocalypse uh was at least in part driven by a perhaps you know naive would be probably harsh but a less sophisticated than it needed to be use of the thumbs up thumbs down signal in chat GBT that fed back into uh post training process and obviously
90:35you know created uh the sick fancy problem. So I've been generally I mean I thought that was a really interesting moment. It showed how much you know we still don't know and the fact that we can be surprised and the fact that we can be surprised post launch in such a meaningful way where I guess you have to put it to a large extent to the differences between how