Adam Ante22:30
Yeah, I remember the day that ChatGPT dropped and I was in an investor meeting and one of the investors brought it up and said, 'Hey, have you seen this yet?' And I hadn't seen it. It was the first day and I remember from that moment though when I started the message that I was like, 'Wow, this is going to be different. This could really change some things.' And I started pushing actually right away back into our product organization. I remember some of the early conversations like, 'Hey, yeah, that's interesting, but it's going to be a little while.' And it has taken a while. Well, it's taken I guess 2 and a half years since that point. And, you know, we've made a ton of progress, but there still hasn't been a ton of B2B like internal process like complete reimagining because of this, but it's definitely been exciting and I think there's a lot of cool areas like research that we've been able to leverage internally, some of the co-pilot type things. But I think you're right, you have to be a little cautious. One of the first things that we did was we stood up an AI council that helped us just think about all of the different items that we should be considered of that we need to be thinking about from a regulatory perspective that we're still, you know, unpacking. How to think about the models and which models to use or maybe which ones to stay away from. And then how do we get it out into the organization effectively and responsibly? How do we start to get it into the product effectively and responsibly? And so that was our first step was really starting with that AI council. One of the things that I've also really gotten into though is to encourage people to try things and do things and, you know, we got to create the right security model around it so that we're not pumping data out into the live web but internally we're using these models and saying hey where can this work? How can we be a little bit more aggressive? Maybe let's try a couple things. So there have been some of those licenses where we tried to get a little bit more aggressive, pilot some stuff, see where we think the opportunity is. And it's one of the reasons why in some cases we say, 'Hey, let's go a little slower here. We're not seeing quite the benefit.' And there are some of the models that software applications came out with really high-priced AI fees right out of the gate. And we said, 'We're not doing that. We don't know what the benefits really could be. Let's pilot it. Let's slow it down.' And then you saw some of those applications actually change their AI pricing strategy afterwards and said, 'All right, let you know, maybe we're not going to get the same fees that we thought we were.' So, we've tried to be really intentional with this combination of piloting, trying, testing things, and then putting some money down on some things to make some progress, but be cautious.