Why the Next Hit AI Product Will Be Social Why the Next Hit AI Product Will Be Social (Best of the Pod)

**Sarah Tavel** (0:00)
Google was a founding team that was deeply, deeply technical. As the technology, the underlying technology, got more mature, the slider goes forward, forward, forward, more towards the product thinker, product experience. Pinterest, where I was, Snap, Instagram, the CEOs weren't technical at all, they were product geniuses. What are the big consumer wins so far in AI? Of course, it's ChatGPT, which is, in a way, not that dissimilar from Google in terms of what it was, just a text box.
Custom GPTs and in ChatGPT feels criminal to me. It's clearly made by a team that is unbelievably capable but isn't social. What's the multiplayer network effect type experience? Someone's going to create a UGC-type community where there are people who are really, really good, make it so much easier for the rest of us.

**Dan Shipper** (1:10)
Sarah, welcome to the show.

**Sarah Tavel** (1:11)
Thanks so much for having me.

**Dan Shipper** (1:13)
So for people who don't know you, you are a partner at Benchmark.

**Sarah Tavel** (1:16)
Yes.

**Dan Shipper** (1:17)
Before that, you were early at Pinterest and before that you studied philosophy, which I also studied philosophy, so that's close to my heart.

**Sarah Tavel** (1:24)
Yes, I've heard your podcast and I was very impressed with your podcast with Reed.

**Dan Shipper** (1:28)
Oh, thank you.

**Sarah Tavel** (1:29)
To keep up with him was a lot of, very impressive.

**Dan Shipper** (1:34)
It was a couple of late nights of me furiously prompting ChatGPT to explain Wittgenstein. I love it. I love it. Well, you did great. Thank you. So I'm psyched to have you on the show. There's so much to talk about, but I think one of your big interests is in consumer technology and consumer technology cycles and how you can use the lessons of previous consumer technology waves to help you understand this AI wave in this cycle and what kinds of products are going to work and what kind of products are not going to work. I'm curious. I think that's a good place to start.

**Sarah Tavel** (2:04)
You know, one thing I just was reflecting on, and you kind of look at ChatGPT and character AI, and I was just puzzling over those, and then started to think back to, you know, what was like the big early kind of consumer web hit? And that was Google. I mean, it was Yahoo and Google. But what was Google? Like Google was a founding team that was deeply, deeply technical.
And really, like, if you think about a product experience that you expose to the user and how much of it is like the UI that you interface with the product itself versus all the magic that happens on the backend and makes something that's really complex, simple on the front end, that was like what Google was, you know, so good at, the distributed engineering, the infrastructure.
And then as the technology, the underlying technology got more mature, you started to go to a place where maybe if you kind of said, like, deep technical, you know, you have 0% and 100%. Like I would say Google was 95% like deeply, deeply technical. And then you start to move that bar over and you get to, you know, I think about Facebook, like Facebook, it wasn't the same technical depth, of course, of Google, but relative to Friendster and MySpace, they were more technical. They had, they were a little bit later and it let them create like a really performant experience that ended up really kind of, you know, winning the day.
And then you progress even further, Pinterest, where I was, Snap, Instagram. The CEOs weren't technical at all. They were product geniuses, right? And so like the slider goes forward, forward, forward, more towards the product thinker, product experience. And then you think about what we, like what are the big consumer wins so far in AI? Of course it's ChatGPT, which is in a way, not that dissimilar from Google in terms of what it was. It's just a text box and character AI, like unbelievable what they did. And it was like a new paradigm, but still it was always like you speak to Noam. And for him, the product was a model, you know, it wasn't, it was a, it was a little bit, maybe 94% backend, but it was still like very much so. So it's still early, like everything is moving under our feet still, and I think to really have the people who have more of that product intuition, to really build the experiences on top, you need more of the underlying infrastructure to be a little bit more stable, but it does seem like we're moving into that next paradigm soon and what's going to happen there.