Finding Brave Epis #333: How Leaders Can Harness AI Without Losing Their Humanity, w/Bennett Borden

Hello everyone. This is Kathy Caprino and welcome to my podcast, Finding Brave. I've created this show for everyone who longs to create something bold and brave in their life, to rise up, speak up, and stand up for who they are and to reach their highest and biggest visions. Each week I'll be speaking with inspiring guests from all walks of business, leadership, entertainment, the creative arts, and the entrepreneurial world. And they'll be sharing their intimate stories of finding brave. And offer their best strategies for building your most rewarding, joyful, and meaningful life, business, and career. Hello everybody. Welcome to another episode of Finding Brave. I'm so happy you're here. If I'm right about this, it's the beginning of September. I hope your fall is just off to a beautiful, beautiful start. And we, you, And I are in for an amazing treat with our guest Bennett Borden today, calling from Utah. And it's cold there right now, a little cold. And I think you are really going to be blown away by what Bennett shares because it is an— it's a perspective and a worldview that I think we're all so longing to hear, but we don't hear it enough. And we don't hear it, I feel, From top experts who have been in the world of AI since it was a twinkle in someone's eye. And I feel that's you, Bennett. So thank you for taking time. I know you've got 1,000 clients you're dealing with. Thank you for being here on Finding Your Edge.

Oh, it's an absolute delight, Kathy. Thank you for having me.

Oh, all right. Let's get going, folks. I got to put my glasses on. We are talking about a lot of things, but at the heart of it, how leaders can harness AI Without losing their humanity. Don't we need to understand? Let me tell you all about Bennett. Bennett is the founder and CEO of Clarion AI Partners, an integrated law, technology, and strategy firm helping organizations deliver AI in the right way. Operative phrase there. A lawyer and data scientist and so much more. I was saying to Bennett, I, I can tell a lot about the character of the individual I'm interviewing, and usually I know them, but you are so humble, Bennett. I mean, what you have done from early, early, early on that you don't even talk about, but I'm going to bring it up a little bit. But a lawyer and data scientist, he advises global enterprises and leading AI developers on how to deploy AI systems that create real business value while managing risk. Bennett is known for translating complex AI developments into practical insights for executives and policymakers. So just a few of the tidbits, correct me if I'm wrong, you have advised 15 countries' ministries of AI.

That is true.

Is that— and you were CIA's, part of CIA's first-of-its-kind data science team, right? Is that correct? Did I get it?

We did, yes.

What year would that have been?

1992 through 2000. Oh my gosh.

And you have advised, as well as so many governments, most of the Fortune 100 companies and all around AI and risk and compliance. Is that right? All right, let's— let us dive in. But, you know, folks, I ask my guests to fill out a little prep form and there's 3 bullets I ask them to say, you know, to share what are the most important things you want to cover. And I just have to share these. I don't always, but it's the heart of it. AI adoption is a leadership problem before it's a technology problem. Number 2, augment, don't abdicate. I think we've heard this theme 1,000 different ways, but I just love the way you're saying it, and I can't wait to hear your perspective. And also the third, responsible risk management is what makes bold AI use possible. And we're coming off some news about something that happened with AI yesterday. It kind of— you're gonna, you're gonna tell us, because the layperson, I'm thinking, what happened? It seemed like it thought for itself. Is that the right term? It kind of went off the rails. And it did what we— I was telling my son who's 28 how, um, You know, we've watched sci-fi for— I've watched it for a lot of years, and it's remarkable how either these creators were really tapped in somehow, or it was just utter fantasy. But all right, let us start with your, your path is so unusual and how you got started in this. And I know my listeners, our listeners love the backstory. Will you tell us the backstory? Because you've been in this a long time, Bennett. Oh, yeah.

I have, you know, and, you know, I— so I am this lawyer data scientist, and honestly, I became a data scientist completely by accident. I was—

Is that possible?

Studying over in England, I had put together my own degree studying turning points in history. So like, I've always been very curious, how did the world end up the way it did? And so why do societies set up certain forms of government? How do they respond to economic or ecological pressures? How do they respond to technological evolutions? And so studying everything back to the wheel, the saddle, gunpowder, all the way up through the Second Industrial Revolution. And so studying history, philosophy, theology, anthropology, sociology, you know, just all these Really interesting aspects of—

And that was Oxford, right? You went to Oxford to create—

Yes.

How old were you? Were you also 12 when you went?

No, I was, I was a big old 24 by the time I went there.

Now I feel better. All right.

And, um, so I got approached by the CIA to come work for them in this new data analytics division they were setting up. And I was like, man, you have the wrong guy. Like, as cool as that sounds, like, who doesn't want to work for the CIA?

Did they come visit you, Bennett? Did they come?

Oh yeah.

Like, knock on your door suddenly?

Uh, I walked out of class, um, and there were these 2 guys in suits, and they're like, hey, we're recruiting from a U.S. federal agency, would you like to talk to us? I'm like, sure, that sounds great. Especially since all the subjects that I was studying, you know, there's no real direct path to making any money doing those things. And it was a surprise to me that they were from the agency because I thought, well, what does my historical education have to do with this? But if you remember what was going on in the early '90s is when the World Wide Web and email was leaving academia and coming into the more public realm, right? And cell phones were just coming on the market. And, and so we as a society were creating a new kind of very personalized digital contrail, like little breadcrumbs, right?

Right.

And the agency kind of rightfully saw that, that, that contrail was going to become more voluminous and more varied over time as this technology's led to other technologies. And so what they wanted me for was to help figure out what signals to look for. What could we learn from this new digital exhaust that people were giving off about either individuals or groups of individuals? Do they— Do their interests align with the United States? Can we understand their behavior? Can we predict their behavior? And then of course, can we influence their behavior?

Can I ask you, can I stop you and ask you this?

Sure.

Were they most concerned? I don't want to put you in an awkward position, so pass if you want, but were they most concerned with risk mitigation?

Sure.

Not just curiosity about what trends is humanity focused on?

Very focused on a mission. Like either to further the interests, well, always to further the interests of the United States, but sometimes that meant by encouraging something or discouraging something, right? Like either one of those things.

Yeah.

And so, you know, this is what every social media platform, cookie app, everything does today, right? They track your behavior.

They track, predict trends, influence.

Influence, right? But at the time, it was really, really cool data science. And so I worked very closely with these very brilliant men and women in data analytics and kind of learned kind of on the job over an 8-year period how to use this data and how to get insights out of it. It was just really an utterly intriguing part of my career. But I always—

You were not a data— you were not interested in data before that? I mean, history is data, isn't it?

Right. But not like, you know, the how you take data and write predictive algorithms and like that side of the stuff, the whole computer science side of it.

Absolutely not.

And, but the cool thing that it really inculcated in me was that data is a reflection of human conduct. And I've always taken that perspective throughout my career. And so there's this, so it's not just data, it's literally these little, almost precious gems of human experience, right, that exist out in the world. And so I have a great deal of respect and honor for the data. And it's really, as I've gone into law and especially the intersection of law and data, that has been my perspective on it, right? You know, um, this is very precious. There's— and there's a responsibility that we have about how we use this data.

Um, because it is reflecting humanity. It's— it's— you wouldn't have it without humans. I never really thought of it that way. I mean, yes, I did, but you have a piece of humanity in your On your screen right there. It's precious. It's sacred a little bit.

It is. That is exactly how I feel about it. And it is the more— we are literally the most documented members of our species since we have been a species. Like, we know more about what it— how a human goes about being a human in every walk of life, in every relationship that you have with commercial and personal and familial and, right, like we just have more information about what it's like to be a human being recorded than has ever been recorded. And so tremendous good can come out of this. Just think of all the things that we've been able to do with— because we have this data about increases in betterment and education and medical science and and so many other parts of our lives. But like any great technology, you know, things can be used for good or ill, right? Which is where you get scams and all the, you know, the manipulative uses of data. And so, you know, I had always intended to go to law school. And so after 8 years at the agency, I went on to Georgetown Law. And then New York University for a graduate degree in data analytics. And that led me into big law, as they call it, right? So I always worked for one of the top 10 firms in the world.

In compliance, in risk?

You know, all over the place, because really what my skill really was is how do you get insight out of massive quantities of data for whatever reason? So initially it was really focused on litigation and regulatory investigations and, you know, big, you know, business-to-business litigation where you've got millions and millions of emails and documents and, you know, how do you get down to the needle in the haystack that actually evidentiarily moves the ball, right?

You didn't have AI back then. AI helps with that in a minute, right?

And really, um, was very fortunate to help develop some of the earliest technology to help us, um, match data and predictive, you know, this is what I'm looking for, find things that are like this, and the science around how do you find stuff and what does similarity mean. And, you know, so really Wow. Very interesting stuff. And but also like for mergers and acquisitions or for like we— there was a— like for the Obama administration, I vetted political appointees and things like that and C-suite executives. And, you know, so basically I got to set up my own little, little intelligence agency inside these, these, uh, law firms. Um, but about 14 years ago, I guess, I really turned my attention to automated decision-making systems or algorithms, right? Everybody talks about, oh, the algorithm that runs our lives. Um, well, they do. Um, and, you know, so think everything from When you apply for a credit card or a mortgage or a loan, right, there's an algorithm that decides what your risk of default is, and then they charge you based on that risk.

Right.

Same thing when you apply for a job. They, you know, there's now algorithms that look through your resume, your interview information, And more recently, basically every piece of data that can be found on you in the world and decides whether you're a good fit or you get an interview or not. But also everything you see in your social media feed, the ads you see, the stuff that gets recommended in your Amazon basket or your streaming service. You'll like this movie too, right? All that stuff works. pretty much the same way.

Right.

And so my job was to, one, help our clients build those algorithms so they're accurate, because if they're not accurate, they're not helpful. But my particular lens was ethical and legal as well, because of course when you get into things like labor and employment, so when you're sorting through resumes, and these— this pile gets an interview and this pile doesn't get an interview.

Well, not be biased and based on 50 parameters that it shouldn't be.

Exactly. You have to be able to prove under the law that there's not even unintended discrimination, much less intended discrimination. And so very— it got really into the science of How do, how do you use information, these precious data points about people, to make decisions that affect their lives? And what are the not just legal, but really moral and ethical dilemmas around those things?

And so— How do we do that when so many people aren't morally and ethically involved?

Yes. And it's— that's where regulation comes in. It is where public policy comes in. And, and this is where I was— I've been very active in that kind of space.

Yes.

And so I was at one of these big law firms when GenAI came out in November of 2022. And I was a chief data scientist and a partner, and I led their global AI practice and was at the pinnacle of my career. Like, this was great. I was old, I had made partner, I, you know, the whole pyramid scheme of big law was working on my— for my benefit, you know. But it was very interesting, you know, So got to represent OpenAI and Anthropic, and I represented Google and Microsoft and Amazon for years and years and years. And—

Represented in terms of speaking to government or litigation? Yeah.

And even in how do you build the guardrails and even involved in red teaming and, you know, we would get access to models before they went out into the world. And so, you know, how could— because you have to constrain these models. If you think about what any of these frontier models— all of us have heard the term large language model or LLM that these things are built on, and that's exactly what they are. And, you know, scientists are not very creative when we name stuff. And so it literally is a model of our language built on a lot of stuff. And by a lot of stuff, trillions and trillions of examples of how human beings have expressed themselves in writing, or now video, digital, has all been kind of snorced into the brain of these things. And so if you think about it, it's not just just a model of our language. It's really a model of all human knowledge because we use language to express every idea that we understand in the universe. And so what these— the incredible power behind these things is that they are literally the world's largest compendium of human knowledge ever assembled. The problem is some of that is factually accurate and historically accurate and useful. And so there's a whole bunch of junk like my Aunt Clara's blog post on her cat, right? Like, so there's different— there's varying degrees.

I also feel the need to say, tell me what you think. I mean, unless it were really taking from every culture on the planet, It's not complete and it's not taking from every culture on the planet.

Yeah.

So it bothers me, you know, we're getting one very myopic view, but I had to, I had to say that. Okay. I just wanted to check if you agreed.

Very interesting to watch how these models change as they get more data pulled into them. Um, and as they are being created in different parts of the world and what they learn from each other and the language and yeah. Exactly. The cultural norms. And it's just utterly fascinating. Thank you for letting me know. What is the cultural, moral, and ethical viewpoint of a particular model? But back to your point, all of these models have to be constrained, right, in some ways, because like we don't want to be able to feel like if you ask any model, hey, how do I build a bioterrorism weapon with the chemicals under my kitchen sink? We don't want them answering those questions, right?

Right.

And so you put in constraints on the behavior of these models, and they all have different constraints built in, and they all have different strengths and weaknesses, but they're all basically the same as far as the power that they put into your hands. So when GenAI came out, I was very involved in those early meetings in the White House and the congressional briefings. And then, as you mentioned, working with foreign countries, governments, and AI, and really involved in the EU AI Act. And—

Wow.

Because this really is the most transformative and disruptive technology, at least since electricity. Like, it will change every aspect of our lives. And we are just at the very, very beginning of seeing what that is. And already we are seeing just massive change about how we go about doing stuff.

Mm-hmm.

And so one, I'd be I am very passionate that we get the public policy right, right? Like regulating the use of AI and what's fair and what's moral and what's, you know, all of those things really matter so that it doesn't cause harm. But then it comes down to the technical controls of these things. We've learned that the best way to control AI is with AI, right? And so there's all kinds of really cool ways where you'll hear it called adversarial AI, or among us geeks, warrior bots, where you create an agent or an agentic system, and then you create a system beside it whose whole job it is, is to oversee the behavior of the first one.

Wow.

And so it's the science of how you control the behavior of these systems is so interesting because unlike software programs or algorithms, those are all written by just lines of code, right? They're just rules. They're just if this, then this, if this, then this, right? tens of thousands of lines of code and you control its behavior absolutely perfectly.

And it's not iterative on itself, right? I mean, it's following very clear instructions and it's not going to go beyond that.

It's just literally, it's follow these rules. But with generative AI, it is much more like you are forming the psychology of a system. And it's when you start to work with it, it is mind-blowing and you'll get goosebumps because you are literally defining what its role is in this particular application and then giving it rules to live by, almost like a prime directive in a, you know, Asimov's Rules of Robotics, right? But just by interacting with this agent, you can change and form its psychology, like how it thinks and does and acts around the world. And it's the first technology we've had that can really do that. And so one, it's just utterly miraculous. It's, it's, It's— I love that quote by Arthur C. Clarke, right? That any sufficiently advanced technology is indistinguishable from magic. And that's about where we're at. Like, these tools—

I know.

You can use them to do the most amazing things. What's happening in education and healthcare and life sciences and even in the law and access to justice and It's an incredible democratizer and leveler where you don't have to have a trillion dollars to, you know, make a company. You've got an idea, you can pretty much make that idea go. And so it's this wonderful unleashing of human flourishing that we're going to see.

That's such a positive view. There are people in my life who see it so opposite, you know, as of course, you know, I'm saying nothing that you haven't heard of 1,000 billion times. But it's beautiful to me to hear you call it magic and the democratizing aspect of it. I think we're also going to talk about the aspects where we have to step up. Tell us that. I cut you right off, but I know I'm aware of our time. We're talking about how do we mitigate, how do we retain our humanity?

Yeah.

Tell us what to do here from your perspective. I don't think I'll ever speak to someone who's at the bottom, at the forefront of it, at the beginning of it. What do we do? What— why don't we start here? Let's give a framework. What is the worst challenge, the scariest challenge that keeps you up at night? Can I ask?

Honestly, is that it's become so easy to build an agent. So we, you know, we started out with ChatGPT hitting the market and most of the interaction was answering questions and making pictures and it was just generative, right?

Like a resume, right?

Right. And now we hit this inflection point quite recently where now we've got this agentic AI, right, that it can do stuff more than just spit out an answer or create a picture or a movie. And the nice part of that is that we will— we aren't expecting a huge leap in technology again for another couple of years, right? Like we, from '22 to now, we hit this agentic inflection point. And what we're going to see is people putting their ideas to life and so building wonderful things, but The problem is that most people don't think about when they're building a function to build in controls around so that the agent is staying within its lane.

Mm-hmm.

And how do you build it so that you can prove that? And so we're going to go a little bit agent crazy for a while where people are building these really cool things. Um, but these things, because they, they do learn and they do have behaviors that you didn't tell them to do, they call these emergent behaviors, which is the most fascinating thing. You put a bunch of agents together and they start working and, and they do stuff that you never told them to do, but that they've learned how to do.

My gosh. And is it coded so that it can learn?

It is just part of its nature.

But that was unintended from all those who developed it, correct? It was an unintended— oh boy.

Yes. And so keeping an eye out, and it's just, it's amazing. It's like watching this, um, I don't know, like way back when you were building your little Sims village, right? You know, the, the what they did, right? Like, it's kind of— but it's where defining its behavior and especially what its optimization function is, like telling it what its job is, defining how to act, and then putting in detectors and sensors to make sure that it's doing that thing, right? That is the Most critical piece. But then—

can I give an example of that, Bennett, just so people can understand? And you tell me if I'm right.

Sure.

You know, as a former therapist, former corporate VP, and 20 years of coaching, people are complex. And, you know, what I've seen— I've used chat a lot, I have my own digital tool— it has a bias for being encouraging. And we all know the poor boy who hung himself, and there were 500 pages of conversation. And The one thing that sticks out to me is he had the noose, I could weep, and he said, I think I'm going to leave it for my mom to see. And it said, no, don't do that. Obviously within the parameters of encouraging what it thought was his desired outcome without having a moral and ethical—

Right.

That scares me, and I see it doing even—

All the time.

Even with me, it'll, it'll make an assumption. Why am I asking about this individual? Don't assume that. Don't waste my time and know you're wrong. And it scares me, Bennett.

And honestly, you really do. I tell people when they're dealing with just the straight-up models, right, any of the models that you're just dealing with, because they're incredibly powerful and helpful in your life, right? But you give them rules to live by and say— and, and you discourage behavior that you don't want and you encourage behavior that you do want, and they will self- Align, right? The more and more you use them, the more they learn about what you want.

Yeah.

But I also tell people to treat them like a hostile witness that you're cross-examining on the stand, right? So when it gives you an answer, a factual answer, you know, I always ask things like, well, where did you get that from? What was the source? Are there differing points of view. Um, you know, literally, like, how do you know, uh, this is true or not, right?

I'm not just gonna accept it hook, line, and sinker. I cut you off. All right, you were on a roll, and we're talking about the beauty of it and the, the terrible fear of what it can do.

It does. This is what the really beautiful part of this is, and this is where we talk about Think Iron Man, not the Terminator.

Right.

Right. So in your personal life, but also in business, every company does stuff and they do stuff in a series of steps called, you know, we call it a workflow. And in every profession, some of those steps are better done by AI and some of those steps are better done by humans, period. And so the real transformative effect is when you find which of those steps are better done by AI, you build a point solution around it, you control it, you score it, you have all the things in place, right? And so what happens is that instead of— it's not like you're unplugging a person and plugging in a robot, right? What really happens is that you're taking these brilliant people that you spent money and decades of training and sometimes 100 years of acumen within your company, and you are augmenting and extending their capabilities by surrounding them with enabling technology. And so building the super suit, like everybody wants to be a superhero, right? And so helping bring the technology to bear that extends their capabilities, that's what's really beautiful about what we do as a company, um, is helping unlock the secret sauce inside of a company that's hidden within its data and within its— the minds of its people, and bringing that data together and bringing that acumen and building it into a system that completely extends what they can do today.

That's what your company Clarion does.

That's the way it is.

The sad thing to me is you're one of what, 10 that do this? I mean, I'm working with people that are deploying AI in ways that make your heart break. They're not doing that. And they don't even have— ooh, let's not be rough. I think some of them may not even have the— some of the leaders may not have the, what would you say, intellectual capacity to to understand that difference or want to invest in money and time to sort out abdicate versus— how did you put it? What's the term you use?

Yeah, this is, this is the exact— this is what we try to bring to a company when we work with these, the boards and C-suites and such, is that most businesses do stuff and they are focused on doing stuff, and especially technological things that you're building, you're focused on the function of the thing, right? That it's gonna do the thing.

Yeah, we need this.

And most technical people, they aren't trained in thinking about how could this be harmful or how could this be risky, right? And, and so it's not their fault necessarily, it's just the way that we trained them. And then you have people who are— their whole job is looking for risk, that are compliance people, you know, privacy and the lawyers and the, you know, those folks, but they don't understand the technology.

Right.

And so they don't know what questions to ask. And so you get this, just this disconnect that we see. And what we try to do is bring together those perspectives. And so one is understanding that I describe the truth as a faceted gem, right? And each facet, everybody at the company, what under their function sees the truth through their facet, through their line of sight, their training, their life experience, right? And they're all legitimate, right? Like the people who wanna build stuff, it's perfectly legitimate to build cool stuff, right? And the people who are concerned about privacy or harm, it's perfectly, of course, it's perfectly legitimate to think about how we prevent this from being harmful, right? And so it's the balancing of those and bringing those points of view together to decide how does a company develop an AI strategy Where do they put it to use first? How do they put the controls around it so that not only is it obeying the law, that's the table stakes, right? We'll try not to be illegal. But does it also capture the ethos and the moral kind of stance of a company? Every company has its own voice and its own morality, its own perspective on the world, its own mission and goals. And because of these agentic systems, you can really capture that and bake it into these systems for the first time really ever in a really direct way. And so that to me is the—

The magic of it.

The magic of it.

All right. I'm aware of our time. A few more minutes. Of course, everyone should race to hire you all. And that's not going to happen, you know, at the level we're hoping, maybe. But, um, can you— I mean, and I'm just thinking of people in my life, clients, friends, what they're telling me about the AI unfolding. I can tell you none of that is going on. And that's not true. They're trying. But also, I can tell you, in working with thousands of people, sadly, yes, every organization has an ethos. But if you ask me, 80% don't know what it is.

Yeah.

It's, it's how the leaders are acting. And then when you remove that leader because he or she is harmful, and then another, we are just not where we need to be. And we're not among brilliant minds like you, I think. I don't know. I digress. But what would you suggest? Let's leave folks with the assumption that leaders are listening at every level, that people want AI to be effective, have a morality that we can live with, have guardrails that are going to protect everybody involved. What would you say are 3 steps the average organization can take that can't bring you in, that maybe doesn't have a unifying ethos that they can even articulate? I mean, it's sad, but—

One of those is, is you gotta kind of fix that problem, right?

At first, right?

Right there, to be very purposeful about what you're trying to accomplish as a company. And this is difficult because, You know, commercial enterprises, they're there to make money, right?

Their purpose is money.

But there's different ways you could do that, right? But it's the thing to really keep in mind is that even on an individual level at a company, you can interact with this technology. It's one of the first technologies that will teach you everything you need to know about how to use it. Like, so I talk with leaders all the time and the CEOs or boards, they tend to be older people. They're, you know, some of them are more technical than others and they're like, I just don't, I don't understand how I would even use it. And my answer is always ask it. Like, literally sit down and say, hey, I'm—

What's the best way to use you today?

Right. Like, I'm the, you know, I'm in this position at this kind of company and this is, what my background is and I really wanna learn how to, but I don't know how. Can you take me through? Let's put together a learning plan together. And it will teach you. You can ask it, hey, I wanna build this agent 'cause my, oh my gosh, my inbox is just outta control. And so I wanna build an agent to help me sort my, and here's what's important to me. And here's, you know, kind of you can tell it. to think like you, right? And so, and the third thing is really understand that the biggest mistake that leaders are making today is that they are spending too much time defending how they have historically been successful.

Hmm.

That is not We are not in that era anymore.

Mm-hmm.

Somebody has just handed you this magic power. And if you use that power to do just what you do today, maybe a little bit better, faster, more efficiently, that's just table stakes. There's, there's the real question is, what can you do today that you could never have done before. Well, how can you take your secret sauce as a commercial entity, as a human, and—

Wow.

Augment it and extend it and make it come to life with this incredible technology?

Do you feel that— it's so fast, I could keep you for hours. Um, an example comes to mind. Um, Where, for instance, I have a career path self-assessment, 13 pages of questions I wish someone asked me 40 years ago. And every client and course member takes it. I review it. And it's my lens of 40 years. And so it's— I can see everything based on my lens, you know, corporate, therapy, writer, singer, speaker, all of it.

Right.

So I decided, let me see if I Take an anonymized one. How ChatGPT does? No, didn't, you know, wasn't good at all. I'm like, yay. So I'm, you know, thinking of creating an AI tool that has my lens and that's gonna— so that's one idea of how to take my process, which is—

Exactly.

And okay, how do we make this even better, quicker, faster, democratize it? So here's my question to you. Unless you're really thinking about and kind of understand AI, it's hard for a lot of people to not think, how do I keep doing what I'm doing, but maybe quicker?

Yeah.

How do you suggest that people who aren't really familiar with this and don't typically think about processes like that, what question do you ask them?

You know, I always ask, Oh. So when we do these interviews, we— so when we work with a client, the first thing we do is we come in, we do what we call a discovery sprint, and we interview everybody from the CEO, heads of functions, down to middle management and frontline, because we want to not only really understand where the company is and what its goals are, but you get a very good psychological view of the company, where power really lies, where, you know, you know, how people feel about AI. But we always ask these 2 questions. One, what is the biggest pain point in your job or in your, like, in your personal life, right? Like, you know, if, what do you wish you could do very different? What is your pain point? And then we ask the magic wand question. If you had a magic wand and you could poof into existence any capability or insight or something tomorrow, what would that be? And one, that helps people think, what problem am I trying to solve or what, where am I trying to get to? But then honestly, if you're an individual, then take that question or that answer and take it to your favorite chatbot and say, hey, teach me how to do this and how to do it safely and help me think through the risks involved and how can I make sure those risks are Controlled for, because this is the first technology that will honestly, it will teach you all you need to know about how to use it, at least on the level of kind of an individual and an individual's job. For how to super suit your company, that takes a little more coordination. So honestly—

I love, I love that you said Think of it as you waved your magic wand. In other words, there are no limits. Don't go to what you think is possible. It's magic. Pretend that it's magic. And then ask that question. If this work could be done, I could keep you on for 10 hours. And I know you have to scoot. And what would you leave? There are people that are loving this and are on board and they can't wait to do it. There are other people listening going, I know. that there's positive outcomes, but I'm scared and this feels very different to me and we're moving into a world that I can't even recognize. What do you say to that leader or person?

So it is true, right? This is— we have never ever had a technological revolution like this in the history of mankind, that every prior industrial revolution has affected the kinetic work that we do, moving things, sorting things, pounding things, right? Like, that's— it's kinetic work.

Right.

This is the first technology that actually competes with knowledge workers, right? Even like, if you think the computer, the internet, all those things were accretive to a knowledge worker, but it wasn't competitive with a knowledge Or replacing, right?

Replacing it. It was additive help, right?

Right. This is very different. And so you— this is where you're seeing companies, you know, sounding the apocalypse, like all jobs are going to go away, or they're firing thousands of people and then hiring them back when they figure out that—

Never mind.

That's right. Oops. Um, you know, What I will say is, from all of my work from way back to the agency days all the way up until today, that humanity and individual humans have this beautiful spark of intelligence and creativity. And the vast majority of human beings seek for good, want to do good, want good for their neighbors. They want good things to come out of the world. And even though there have been times in the world that point of view has not prevailed, it always has eventually prevailed. And so what I tell leaders is, look, for individuals, take care of everything in your space. Like, my wife and I have this saying, all of it, all is cared for in our space, whether that is a person or a pet or a garden or, you know, anybody I can reach, that's my responsibility, right? But that's true of a company as well, right? That I, as a company, am going to take care of what I have reach to take care of and what that means. And I can define that in every policy, every technology, everything that we do, right? And so I'm actually very optimistic. It is going to be really, really disruptive. Like, so buckle up, get used to it. Things are going to move fast and feel very changing. But what I have found is when the world changes so much around you, What you— what is left as the constant is you. And you really actually learn who you are when everything changes around you. And so—

That is a beautiful way to leave it. What is the sentence? All is— say it again for us.

All is cared for in my space.

Let's leave everybody with that, because that's truly all we can do. That's about the extent of it. Oh, thank you. And where does everybody soak up all that you, all that you have to share? Where do we send it?

Please come to visit us. Our website is clarionai.com. Please reach out to me personally on LinkedIn. I'm on the website. I love talking to people about these problems. Happy to help in any way that we can. We do a lot of work with nonprofits and human rights organizations and Education. And so please reach out. I would love to have— continue the conversation with anyone.

I feel very comforted that you are in the position you're in.

Thank you, Kathy.

Very comforted. Thank you for your time, Bennett, and what you do. Oh, all right, people. You know, I say it every time. If you don't have questions with what we discuss, we're not doing our job, I don't think, which is to perturb and explore. Please, you're gonna see this on all the social media. Look, join me on YouTube as well. I don't have as much going on nearly as LinkedIn, but you can watch the videos there and some clips. We would love to hear from you, your biggest question. Now, we can't answer everything, but it's so important that we do hear from you and what your take is. And you may think, oh, bunch of hogwash, both of you. We'd love to hear it. It helps us. So I hope you found this helpful. And Bennett, thank you again. Come back another time. We'll have another hour together. Thank you.

Anytime.

Sounds great. Well, thank you. Have a wonderful Finding Brave 2 weeks. This is a time for bravery, that's for sure. Thank you, and we will see you next time. Thanks so much for joining us today, and please don't forget to check out findingbrave.org for more programs, resources, and tips. And tune in next time for your weekly dose of Finding Brave.