Why AI Needs Leadership

In today’s episode, I sit down with Noelle Russell, founder of the AI Leadership Institute, bestselling author, and expert in responsible AI, to talk about what it takes to build AI systems that actually work for your business. Noelle explains why leaders need to take ownership of how AI operates, from defining the human role to creating clear instructions, examples, and constraints. We discuss why treating AI like a vending machine creates problems, how easily poorly designed systems can go off course, and why responsible AI ultimately comes down to leadership, accountability, and staying actively involved in the process. See Privacy Policy at https://art19.com/privacy and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info.

• 0:00Speaker 00

  Welcome to Office Hours. I'm David Meltzer here, live at the Sirius XM Studios at the Wynn in Las Vegas. Oh, I adore this next guest. Her name is Noelle Russell. She's written the book Scaling Responsible AI. But she is an amazing entrepreneur, engineer, creative innovator, TEDx Award-winning speaker, if that's not enough, a bestselling author as well, the leader herself, AI Leadership Institute. That's a lot to carry on your shoulders. Welcome. And she has six kids, married to this wonderful man here that you can't see on screen. That's incredible. Welcome, Noelle Russell, to Office Hours.

• 0:39Speaker 01

  Yay, thank you so much. The weather's fine over here.

• 0:43Speaker 00

  So far, so good. People tell me, like, you know, I only have time to do this. I think.

• 0:48Speaker 01

  A hundred percent. Yes, exactly.

• 0:50Speaker 00

  So, quick, quick question for me. What is the type?

• 0:56Speaker 01

  Yes, this is my favorite analogy. I actually was a, so many of you know I was at Amazon Alexa. But after Alexa, I went to Microsoft and my job was to manage seventeen research teams. And I used to say it was like herding cats, but I actually uncovered it was more like herding baby tigers. Because these baby tigers, like little AI systems, every time you use an AI, you're like, oh, my gosh, this is pretty cool. Oh, my gosh, this is really nice. It's enthusiastic, fluffy, cool, nice to use. But at that moment is when you have to ask the most important questions, things like, you know, look at its paws. How big is this thing gonna be? Or razor-sharp teeth, right? What is it gonna eat? How much is it gonna eat? Where is it gonna live? Who's gonna take care of it? These are some of the most important questions that I think need to be asked, and you can't ask them at the end.

• 1:41Speaker 01

  It's so important to ask them at the end.

• 1:44Speaker 00

  Wow, that's a lot of responsibility.

• 1:46Speaker 01

  Yes, indeed.

• 1:48Speaker 02

  Yeah, so, Noelle, we connected with a lot of love and enthusiasm. And so, I love AI, so of course we're bringing it into our workforce. And I have a digital twin AI showing up in a couple of weeks. I'm so excited about. But what I really love is that you are.

• 2:08Speaker 00

  There are DNA diamonds in your life.

• 2:10Speaker 02

  Yes, there are.

• 2:12Speaker 00

  Copy.

• 2:13Speaker 02

  Yes. So, helping business owners and creators hiring six-agent AI workforce, right?

• 2:22Speaker 02

  That's the direction and the future we're going for. How can we do that responsible ourselves as entrepreneurs?

• 2:30Speaker 01

  Absolutely. I think one of the most, so one of the things that I focus on is the pain that founders are in right now, that entrepreneurs are in, right? Being drowning in hundreds of emails that aren't answered. Last year, I lost $210,000 in a speaking opportunity because I didn't get the email. It just didn't surface. I found it later, and that hurts more. I wish I had never seen it. But that's the reality, right? Not getting on social enough, not being consistent. The algorithmic heartbeat is required, and every founder needs to do it. But it's exhausting, and most of us are just trying to build a plane and fly a plane and like all at the same time. You know, as a mom of six kids, I get it. And so, responsibility actually has become so important to me

• 3:14Speaker 01

  because if you are a founder and you start using AI, your tendency, baby tiger mode, is you're like, oh, my gosh, this is awesome. I mean, I just talked to somebody, put something out into the world. Nobody tested it, no guardrails, no policy. Put it out into the world, and the AI failed them. And 95% of companies have AI that's failing them right now. So, my job is to make sure that your AI is secure, accurate, trustworthy. And so, that's how you do it. You become a better leader, which is why I love the work that we do, all the amazing people on this show, is because we're talking about what it takes to be a good human leader, and in the world of non-human identities, it's the exact same skillset.

• 3:55Speaker 02

  So powerful. Thank you for sharing that. Thank you.

• 4:00Speaker 03

  I mean, I have so many questions.

• 4:03Speaker 03

  Where do I, Avram, begin? But I, I think that, you know, you've built this amazing career at the intersection of innovation and responsibility. I spend my career on identity, rebuilding it, and I, my question is, what AI has taught you about human identity? And maybe what have you learned from it in protecting it? Because it's very easy for us to elaborate on how great we are via AI, but we want to make sure our identity is intact, so.

• 4:35Speaker 01

  Absolutely, yeah. So, one of the things that I've learned over the years, and I, of course, implemented it at Alexa, but we didn't really know what we were doing back then. When I got to Microsoft, I knew things I didn't want to repeat. But one of the things that we've now found out is that there is a system for building AI that works and scales. And that system is called an AI Safety System, and the top brands in the world, they build this way. And the first, it's a bunch of concentric circles, but the biggest concentric circle is human AI experience, that you, as the leader, as the person building the system, designing the system, you have an idea and you want to go to market, and you're like, I mean, we just had a guest, right, that was talking about they are going to build, you know,

• 5:19Speaker 01

  something that they want to get to market, and they leveraged AI to help them build out that idea. But the most important thing to do is think, okay, what is the AI gonna do for me and what is the human gonna do, right? Who am I and how am I gonna stay present in the process? And most companies, overprivileged, they, they're like, AI do all the things, talk to my customers, reply to my customers, close my deals. When the reality is, is that there are certain things as a human, we need to own, and that's the ultimate responsibility.

• 5:47Speaker 03

  I agree. I agree.

• 5:48Speaker 01

  Taking ownership.

• 5:49Speaker 03

  Yep. Thank you very much.

• 5:50Speaker 01

  Yeah, you're welcome. Thank you.

• 5:53Speaker 04

  When it's time to scale your business, it's time for Shopify. Get everything you need to grow the way you want, like all the way.

• 6:03Speaker 04

  Stack more sales with the best converting checkout on the planet. Track your cha-chings from every channel right in one spot. And turn real-time reporting into big-time opportunities. Take your business to a whole new level. Switch to Shopify. Start your free trial today.

• 6:21Speaker 03

  This has been an enlightening conversation. I wonder, what does it mean, scaling responsible AI? And then, who's responsible? Who's responsible for actually making sure we're using it correctly and appropriate?

• 6:32Speaker 01

  Yes, you know, sad story, you. And I don't mean you per se, but all of you.

• 6:39Speaker 00

  The sad story is you.

• 6:41Speaker 01

  Congratulations.

• 6:44Speaker 00

  It came across a little weird on TV. Sad story.

• 6:50Speaker 01

  Congratulations.

• 6:51Speaker 00

  I'm his business partner.

• 6:54Speaker 01

  But I speak on stages all over the world. Last year, I did ninety events, tens of thousands, actually, over a hundred thousand people. And in every one of these stages, people are thinking, AI is out there, AI is this big thing, the AI in the sky, Skynet, right? This idea that AI is going to do something to them. And what I have really hoped to encourage them to think, I always say, I'm like, I invite you to think a little bit differently about this. And that's, many of us use AI like a consumer, right? We go to AI and we're like, give me an email, or help me write a LinkedIn post, or build me a business plan, when it's like vending machine AI. It's the worst form of usage.

• 7:36Speaker 01

  Because every single system, and this is like, good news and bad, every single system has been trained by someone like me, likely, a data scientist. Every system's been trained, every system has instructions, every system's been told what to do. But if you're using an AI and you didn't write those instructions, someone else did. And they wrote it for a different problem, for a different solution, for a different audience, and you're leveraging it as if it's truth. And that is probably the biggest challenge most companies run into, is that they default to a system that wasn't written for them. And they don't realize that it's not a technical skill anymore. Like, all you have to do is take ownership. So, when we're talking about that safety system, the first one's human AI experience. What's the human gonna do? What's the AI gonna do?

• 8:18Speaker 01

  The very next thing is system prompting, moving from being a consumer of AI to being the producer of AI, knowing that when you build a system, you can tell it what role to play, what task to do. You can give it examples of what good looks like and bad looks like, and then you can give it constraints, which is the number one thing organizations miss. It's the number one reason. I go in, I help a dealership. I have a lot of auto dealers as clients. I go help a dealership. And one dealer, we did a test, and the AI sold a Chevrolet, a Chevrolet Tahoe for a dollar. The AI agent sold it, and it was a good reason. You go look at the transcript, and the person's like, I can't really afford it. And, and I really, and, and the funny thing is, we told the model, do what's right for the customer, right?

• 9:03Speaker 01

  Don't we always, we tell our salespeople, you know, we don't tell them the customer's always right, we know that that can go sideways. But we do say, like, do what's right for the customer, like, make sure you maintain that relationship. And the conversation took a turn, and now, wait, wait, wait, we didn't mean sell a car for a dollar. We didn't mean, you know, like, give away free product, right? Like, influencers are running into a situation where they're having AI run their business, and AI is coming in and saying, oh, my gosh, somebody asked for our most expensive product, and they just want to try it. And the AI, if not told not to, will be like, absolutely, we want to be partners. This is the best way to create partnership. The only thing that's the difference between AI gone wrong and AI that's profitable and safe

• 9:45Speaker 01

  is a leader that knows how to tell the machine what to do, that is disciplined enough, that has my best self-awareness, right? That you even, same things you tell yourself, you have the courage to tell a machine. And most people don't, don't know how to do that.

• 9:58Speaker 03

  Sounds like how my seven-year-old son would run a business.

• 10:00Speaker 01

  Yeah, yeah, exactly. At scale.

• 10:04Speaker 00

  When we talk about responsible, asking ourselves, what did we do to be accountable or responsible for this? The bigger question is, what are we learning from it? And it's now, today, not only what we're learning from it, it's what AI is learning from it as well. And there is a component or an aspect of artificial intelligence that requires an omniscience of ability to see the future. And as we speed up the processing of wisdom and definitely accessibility to knowledge, there comes time when we need the right people to help us, to empower us.

• 10:42Speaker 00

  And that's what AI Leadership Institute does, is it helps leaders utilize AI to protect us from baby tigers that, as we've seen in many movies.

• 10:54Speaker 01

  Go wild.

• 10:54Speaker 00

  Go wild, and not just girls gone wild, baby tigers. So, we need leadership and experts out there, and I'm glad that more stages are being given to Noelle Russell because the world needs to learn about how to be responsible. Because I do believe that's everyone's intent through the workflow process of creating AI and the agents that are created beyond. We are blessed to have our friend in our neighborhood, Noelle Russell, empowering others, doing power others, to be happy, to make a lot of money for the sake of helping others and having fun. Noelle Russell, it's an honor and a privilege to have you on Office Hours.

• 11:34Speaker 01

  Thanks so much. Just call me your new AI BFF.

• 11:38Speaker 00

  New AI BFF in our neighborhood here on Office Hours.