Why Most AI Use Cases Miss the Point | Taylor Black (Microsoft)

**Les** (0:08)
Data Storytellers, long-awaited episode here on the show. We overcame all kinds of obstacles and difficulties to be here, but it's my pleasure to welcome on the show, Taylor Black. Taylor, good to have you.

**Taylor Black** (0:23)
Les, thanks for having me. I'm excited for this.

**Les** (0:26)
All right, so today we will be ambitious. We will try to cover a lot. I mean, obviously we are wrestling with the topic of AI, which is a challenge.
It's moving fast. There's a lot of attention going into the topic, a lot of money pouring in, and there are all kinds of takes from all sides of the aisle. You know, people are dropping their two cents. So we will try and stay grounded in reality today. So what I would like to do is really to... I mean, every good story has a beginning and a middle and an end, right? So maybe we can start with your perspective. So, first of all, you are right now the director of the AI Ventures ecosystem over at Microsoft in the CTO's office, right? It sounds very exciting. To be honest with you, I don't know exactly what it means. I have an idea, you know, based on the research I've done and the conversation that you and I had, but I intentionally left some things still in the... they're shrouded with mystery so that we can uncover that today. So would you mind shining a light on what that role actually is?

**Taylor Black** (1:35)
Yeah, for sure. Right, it's a proper large company title where it doesn't exactly tell you what's being done there. But I think the key thing is that I'm in the CTO's office. The CTO's office is an extension of the CTO. What that means is we kind of do whatever Kevin Scott needs in terms of getting Microsoft work done. My niche, tiny niche of that ends up being working across the company in early-stage venture building sorts of motions.
And also kind of working outside of the company to understand what sorts of signals and insights we can gain from the broader early-stage startup ecosystem. And as part of that too, I run an internal peer-reviewed conference motion in the AI ML and data science space where we have a huge set of folks internally, about 23,000 or so practitioners of this technology in really deep ways, where we convene them about once a month around a peer-reviewed conference. It really helps our knowledge sharing and our excellence internally, to be able to talk about Microsoft confidential sorts of things internally with our colleagues. And it's been a really great practice that Microsoft has developed over the last 12 years or so that I've recently become a part of as well.

**Les** (2:59)
So context is everything, and I would like to explore what you actually do and how to translate into other businesses as well. You know, a lot of our members, actually all of our members are big enterprise data analytics and AI leaders from Fortune 500 companies, and they would love to hear from you and your experience of how you drive this internal innovation engine over at Microsoft. But I think it's important to understand how you even ended up in this role. So with regards to your background, maybe we can go all the way back as far as we need to. Like, how did you get into data and analytics and AI?

**Taylor Black** (3:38)
Yeah, totally. Well, I mean, if you want to go all the way back, it was in my parents' basement in the late 90s, learning how to program. I kind of taught myself programming in high school, and that ended up being a great way of paying the bills all the way through undergrad, grad school, and law school. Slowly developed into being, you know, quote unquote, a full stack dev in the web space, building websites and web apps and mobile apps for, as a design and development firm. Discovered we started building the same sort of thing over and over again and turned it into a B2B SaaS product, a simple kind of learning management system right when those were really becoming all the rage. You can kind of think of it as an earlier, less great version of Corsera perhaps. But it did the trick for our customers and provided a lot of value for our customers. And so we learned a lot about a lot about that. Where data really came in though, it was right about the time when companies like HubSpot were spinning up when Google Analytics was really making its penetration into the market. And we discovered that we could easily sell our solutions to our customers because we were able to show them an ROI before we even built the thing as a result of using Google Analytics for SEO performance, for conversion performance on their actual website as a result of the way in which we built things. And our jobs became dramatically easier in like a three-month period because we didn't have to sell our customers on a better design or a better widget. We just showed them the conversion rates that they could get concretely as a result of building their website in a certain way and following certain search engine optimization processes.