The #1 Mistake AI Startup Founders Make — You.com's CPO, Saurabh Sharma, Reveals the Truth

AI startups are booming, but many are going to fail. If you were mentoring a founder who is building an AI today, what's the biggest mistake you would warn them against?

Yeah, great question. I see a lot of startups and the ones that are most compelling to me is the ones that are looking a bit over the horizon, if you can. It's something that is right in front of you. It's likely that somebody is already doing it. There's probably five or 10 other companies already doing it, and they're going to be a bit ahead of you. And so the startups that I tend to see as being the most likely to succeed are the ones that are thinking a little bit over the horizon. And then the question becomes, well, how do you think over the horizon? And what I end up seeing is that startups that are focusing now, if you're a verticalized startup focusing on a workflow that is difficult and hard, and nobody seems to kind of have a great idea on how to do it, start working on that. Because what's likely going to happen is that the models are going to get better and better and they're going to enable your use case in a short amount of time. But being there, going deep with those customers now, understanding what can be built now, is going to give you a leg up so that when the models catch up, you're there. You've got to look over the horizon. That's my number one feedback, and that's the number one thing I talk to founders about, is if you know how to build it today, somebody's probably already doing it. So you might as well pick a problem that looks really hard, that looks untractable, start to break it apart and expect the models to catch up to what you need.