Ethics at Scale: Navigating AI Risk | Reid Blackman | 648

And welcome, welcome, welcome. This is Peter Winick. I'm the founder and CEO at Thought Leadership Leverage, and you're joining us on the podcast, which is Leveraging Thought Leadership. Today, my guest is Reid Blackman. He's the founder and CEO of Virtue, which is an AI ethical risk consultancy. He's the author of Ethical Machines, which just came out fairly recently by HBR Press. And he's the host of the podcast by the same name. So without getting into all the nitty gritty and the amazing companies he's worked with and articles he's written, I'd rather just talk to him 'cause he's sitting right here. So welcome, Reid. How are you?

So far so good, but you know, the podcast just started. We'll see.

There we go. That's encouraging. There we go. So what is, uh, I, in terms of categories of jobs that didn't exist 10 years ago.

Yeah, seriously.

What does an AI ethical risk consultancy CEO do on a typical day?

On a typical day? Well, I mean, most of the work consists in, if I'm not, you know, writing and that sort of thing, uh, advising mostly senior executives on how to design what I call an AI ethical risk program. They often call them responsible AI programs, but the bottom line is, all right, look, we're this large organization, we're a Fortune 500. We've come to the point at which AI is coming through the cracks. You know, it's coming through, it's coming through HR, it's coming through in-house development, it's coming through marketing, et cetera, et cetera. How do we, how do we get our hands around this thing so that we can, we can manage the risks of it?

Got it. Because I mean, one of the interesting things that you said there, it's not just, well, you know, make this the CIO's problem 'cause everybody's touching it in some way, shape, or form. And if they're not today, certainly they will be. Very soon.

Yeah. It's fine for the CIO to own the program.

Yeah.

Own the AI. But, but yes, it's, you have to think a lot about, we're not just designing it in-house. It's, you know, every startup wants to sell their new AI solution to every department, every role in your organization. So how do we get our hands around that? So I sort of, I think that in a lot of cases, AI sort of acts like a Trojan horse because HR brings it in and before you know it, you're facing down a lawsuit for discriminatory hiring practices that were the result of your AI being used in a certain way.

Right. So gimme a little bit about your path, 'cause I would imagine that a young Reid didn't sit around saying, what do you wanna be when you grow up? And it was AI ethical risk consultancy.

Nope.

There was probably—

yeah. Yeah. No, I did not. I didn't quite see it coming. So no, young Reid became obsessed with philosophy and, you know, senior year of college, it was either law school or go get a PhD in philosophy. And I realized it'd basically be a betrayal of my identity to, to stop doing philosophy at this point. So I went off and I got a PhD in philosophy and then I became a professor. I was a professor for 10 years. I always specialized in ethics. That's always, there's multiple branches of philosophy. That was always the most interesting to me. And I can get into details if you like, but at some point it occurred to me that I could start an, a, I could start an ethics consultancy of some kind. But I didn't see any kind of market for that sort of thing. And you fast forward a few more years, maybe even more, and I was always keeping sort of, you know, my eyes open. And that's when I heard things like MeToo and Black Lives Matter and #deleteUber and #deleteFacebook or boycott whatever. And I thought, okay, if, if companies don't get their ethical houses in order, that's going to be really bad for them on social media and ultimately news media. I could sell that. And so there's lots more details to that, of course, and lots more behind my motivation, but that's how I ultimately came to start this company. It's because I realized, oh, things are going sideways. I, and also, I don't know how exactly, but maybe it was sort of in the academic ether that AI was coming and things were gonna go sideways with it. And so that was also part of the vision when I started the business is I, I wanna really focus in on, on AI at some point when that becomes sufficiently relevant to the marketplace.

Right. So you saw it early. So it's this intersection of philosophy and tech, but then, you know, you sort of fast forward through, oh, then I was a professor for 10 years. Most people then don't leave academia to take the risk to be an entrepreneur. They say, I hired a professor after 38 years.

Right.

Live happily ever after.

And yeah, sure.

You know, whatever.

Sure. So why, why did you decide to leave the confines of academia? Confines is the right word. Uh, so there, there, there's, there's a lot to say here. I mean, one thing that I glossed over is that when I was in grad school, I started a fireworks wholesaling company based outta Connecticut. And wait, wait, you're, you're, you're at a crossing the road.

Do I become a philosopher or an attorney? But I'll be in somewhat of an arms dealer selling fireworks.

Arms dealer's a bit strong, but, uh, yeah, you know, as a grad student I was making bupkis and my father, The bupkis thing is a technical term. It is a technical term. Yeah. Yeah. Bupkis, Inc. And, uh, so there's why, how did I come to, I started a fireworks business in particular, also part of a story, but my dad was a wholesaler, is a wholesaler. Anyway, so I started selling fireworks outta my two-door Honda going sort of store to store. Like the trunk was crammed with fireworks and the, the back seat and the passenger seat was crammed with fireworks. And, uh, My dad told me where he doesn't have any customers. This was in Connecticut. And then I just sort of drove around store to store selling my wares. And that went really well the first year and went really, went even better the second year. So by the third year, you know, I'm still in grad school. I, I, uh, I had, I rented a cargo van and I crammed that full of fireworks. Anyway, the business grew and each year to my surprise. And so, you know, by the time I had a PhD, I, you know, at some point I had a PhD. And I was driving around in a van in a tank top, listening to Led Zeppelin, delivering fireworks. And I was like, okay, this is, this has gotta, this has gotta stop. Why don't I hire someone to do my stuff and then I'll go off to a coffee shop, I'll do my writing, I'll do my research. Yeah. Whatever. And then I'll check them in. And again, to make a long story short, 'cause we've only got like 20 minutes total here, that's when I learned that managing people is a lot of work because I had 4 or 5 people working in the warehouse pulling orders. Yeah. And loading trucks. I had 4 guys on the road making deliveries. And so I was doing a bigger business, but I was, did not have the free time that I anticipated. But this is what planted the seed of, first of all, that was sort of obviously entrepreneurial. And that's what explains why once I was a professor during those 10 years, I became an advisor, an advisor to startups as part of the entrepreneurship organization at the university. And I saw all these kids doing cool new interesting stuff and I was like, I wanna do something cool, new, and interesting. What would that be? Well, I love philosophy, I love ethics. I don't wanna leave any of that. I wonder if there's such a thing as an ethics consultancy. And I looked around, I Googled it, I couldn't find anything outside of something, you know, very narrow stuff in healthcare. Yeah. And so I thought, okay, there's nothing here. I'm just gonna wait and maybe something comes along and maybe, maybe there's one day there's a market for it and maybe there's not. And then we get back to Black Lives Matter, delete Facebook, blah, blah, blah, blah. Yeah. Cambridge Analytics scandal. That was a big one. Yeah. Um, this was also around the time when all the big consultancies were coming out with these reports about how millennials want more meaningful work and they'll take lower pay for higher meaning.

Yeah.

That sort of thing.

Yep.

And so that's, you know, and then I also, I was a professor at Colgate University, sort of the middle of nowhere.

Right.

My wife and I are city folk, not country folk. We had two kids on the way. We had one kid already had arrived and there, there's another one on the way and sort of back and forth. Camille, yes, my wife's career is Manhattan-based, that didn't work anymore. And anyway, so you sort of put these social factors together with my finally seeing at least what I believed to be an opportunity. And so I left academia and I started Virtue.

So this is where things make total sense in the rearview mirror for thought leaders where you've got, you know, if I were to sort of dissect the smoothie that is you, it's like, you know, 10% entrepreneur, 50% philosopher, you know, whatever. And, and all those shit come together, so you've basically created the perfect situation, the perfect job that only you're suited, or probably only you were suited to have.

I mean, I don't wanna be particularly braggadocious, but yeah, I mean, you know, academics, academics, I'm not, I'm not a nor— I wasn't, I was never a normal academic. I mean, you know, my grad student friends were not selling fireworks, you know, making deals with, making deals with oil companies to sell fireworks to all their gas stations. You know, the convenience stores that they're at. Like, they don't have that business experience. They're not that savvy. You know, I grew up with half my family being from Brooklyn. You know, there's just a level of sort of street savvy that, that, uh, most academics don't have. They don't have experience running a business. Yeah. So I sort of, you know, in a pretty unique position. Plus my wife was sort of in a, you know, at the time she was running a, she was running a business, not her own, but someone else's. She has friends. So I had sort of people to talk to about it. So yeah, it, it all worked out.

And now you're getting, you know, paid to solve these problems that really haven't been solved quite the way they need to be solved before. Because yeah, you know, corporate ethics has been a thing for, let's say, you know, a quite a long time.

Yeah, yeah, sure.

That AI piece and, and the scale and the speed and the intensity at which good or bad things could happen in terms of reputationally, brand and all that, it's huge. If you're enjoying this episode of Leveraging Thought Leadership, please make sure to subscribe. If you'd like to help spread the word about our podcast, please leave a 5-star review at ratethispodcast.com/LTL and share it with your friends. We're available on Apple Podcasts and on all major listening apps. As well as at thoughtleadershipleverage.com/podcast. So who inside of an organization, 'cause I'm assuming Fortune 1000 kind of thing, is ultimately responsible for hiring you or not? 'Cause I think a lot of thought leaders struggle with, hey, here's a problem, I'm uniquely qualified to fix it. Uh-oh. When I go knocking the doors at Citibank or IBM or, or whoever, Who am I, whose door am I knocking on?

Yeah. Okay. So I'm going to say two things. One, you didn't ask a question, but I'm going to say yes, that's exactly right about AI making everything more complicated and we've had corporate ethics, but the truth is that I wasn't particularly interested in doing Black Lives Matter or DEI, whatever, or those non-tech ethics consulting early on. I just thought that's where the only market is right now. And I hope that the AI ethics stuff becomes relevant, that there's a market for it because That's way more intellectually interesting to me for the exact reasons that you specified. It's so fast moving, it's complicated. They're novel problems. So the fact that it's sort of a, a novel puzzle to solve made it intellectually satisfying and stimulating for me as opposed to don't sexually harass women. We know the answer here. Like, that's not, it's not complicated.

Okay.

It's not an open-ended question, right? Is that right? Just don't do it. It's not. Now you have to get people to stop doing it, but that's not intellectually challenging. I don't think it's just some, it's just, you know, well, but stay there a minute because I think, yeah.

This is an interesting piece in, in many instances, thought leaders have their models, their methods, their framework, you know, a change management model, method, framework. And the thinking is, well, it kind of doesn't matter what change your organization is going through. I've got my model. I'm going to impose it on you and it'll probably do better than most at getting you what you need. I would imagine your world, and, and maybe I'm wrong, is you do have your models, your methods, your frameworks, your research, et cetera. But it's not like there's been tens of thousands of cases to go against and 50 years of history.

Yeah.

Is it agile or how much of it is evergreen and how much of it is bespoke?

Uh, okay. Uh, okay. So it, it's very agile. Everything we do is bespoke because organizations are constructed in such different ways. I never went in with like a framework, something like that. Frankly, I barely have a grip on what the hell people mean when they say they've got a framework. Like that usually means they've got a pie chart with 4 buzzwords in it or something along those lines. Uh, and I just think honestly, to, to, to sort of be, I'll say the controversial thing, which is I kind of regard frameworks as ways that experts try to convey things to people who don't know a damn thing. Like that, that's it. It's here. Think about, think of it this way. You're a master expert surgeon, and when you go into a surgery, what's the framework that they She doesn't say, oh, I've got this framework. It's, you know, it's got 4 letters. I've got an acronym for how to do heart surgery.

No, no, no.

She cuts open, she assesses the situation. Of course she's done this a million times before. She's already, but she sees what needs to be done. She sees, you know, what, what's the highest priority here? How do I, how do I tackle that? But she's not sort of ham-handedly, ham-fistedly. I don't know. There's, there's some body part involved in ham. Hand for done. You know, take this framework and apply it to a heart, this particular heart surgery, heart surgery. You know, if you were, if you had a beginner heart surgeon and they come in and they're like, okay, I cut this person open, let me consult my framework. You'd be like, holy shit, this person is gonna perform heart surgery. So I think frameworks frankly are kind of dumbed down versions of what the experts know. It, they're, they're ways of communicating some expert level information to non-experts.

They're valuable as they're valuable on a couple levels, but as a communication tool to convey to others, we use change management again.

Here's 3% cycle.

Here's where we like to be. This is the part that's— so it's actually for you to feel stressed or whatever at this phase, but there's light on the end of the tunnel. Yeah. Right. Value as a communication tool.

100%. Not saying it's not valuable, but what I'm saying is when you asked, you're an expert, do you come with your framework? Right.

No, no, no, no, no.

I come with deep experience and understanding about what's, what's going on. I need to assess the situation and see, okay. Who's actually involved in this conversation when we're talking about AI ethical risk or responsible AI, whatever you wanna call it? What problem are they trying to solve? Do they have the buy-in that they need? What are they really trying? Are they just trying to skate by regulatory compliance? Are they trying to actually go above and beyond that? So there's a bunch of stuff I need to figure out about them before I say, okay, so here's where I think we should go. Maybe, maybe that we need a workshop to get the internal alignment that we need because most companies don't have it.

Yeah.

So let me ask you this then.

So oftentimes with thought leaders, there's sort of two situations, right? Situation one is it would be wise for this organization to learn about dot, dot, dot ethics and AI and all that. And that would be a good thing for us to do because, you know, that market's moving quickly and, you know, it's a capability development and it's thought of as such inside the organization. Let's, teach these leaders or managers the basics of AI and ethics, and we'll pay a certain price for that. Then there's the other side where it's more triage, right? Like when do people buy burglar alarms? The day after their home was robbed, right? Not 2 years before or whatever.

Yeah.

So are you doing more triage or prophylactic kind of preventive capabilities development?

I'd say it's primarily prophylactic. I agree with you. I mean, I think— Totally. Things go sideways with AI, then they're calling their PR firm. You know, they're calling Edelman. They're not calling me.

Yeah.

Because I'm go— I'm not going to sort of tell you how to fix your image or something along those lines. I'm gonna tell you how to do it right. I'm gonna, I'm gonna try to, I'm gonna tell you how moving forward, this is how you do things.

So it's a PR optics problem when they're calling an Edelman, not a get to the future, not happen again.

Yeah, that's right.

A root cause guy then.

Yes. And I think that's why I get to work with these big brands, right? Because the big brands have brand reputation to lose.

Yeah.

You know, some small, medium-sized business, things go sideways for AI, with AI for them at, you know, maybe they face a lawsuit, in which case they're gonna call their lawyer, again, not me. But these big brands, they want to use AI in a big way and they know that the journalists are looking at them, social media's looking at them.

Okay.

And so they better get it right.

Interesting. Interesting. And any trends that you're seeing in terms of it's more likely for a CPG to call you than a professional services firm, or what are you seeing in, in the marketplace in real time?

I work with several CPGs. What I found is that I think that people don't want to design a seatbelt if they don't know if the car works. Right. So They're designing cars right now and they're like, should I invest in AI ethics or responsibility? I don't even, let's see if the damn car moves from point A to point B. And does it move from point A to point B fast enough? Is there an ROI on it? If yes, okay, let's design a seatbelt. Now, who has the ability to see if the car can move? Healthcare is nervous as hell. Why? Because you're literally talking about life and death. CPG, you're talking about ketchup sales, chocolate sales, right? So. CPG is moving faster in AI adoption because they're in a lower risk environment to begin with, and they are therefore also moving faster in the response to AI ethics space.

Got it.

So I would've thought, and I would, and I was wrong, I did think, and I was wrong, that healthcare would be really early on because healthcare speaks the language of ethics. They've been doing it for decades. They have institutional review boards, ethical ethics committees. Yeah. Yeah. IRBs. Right. But they've been the slowest and I think they've been the slowest precisely because they've been the slowest to adopt and they've been the slowest to adopt because they're in such a high-risk situation.

Yeah. Well, and then, and then I would, I would imagine that there's tech that has to, so when you talk about autonomous driving and, you know, the trolley car dilemma and all that sort of stuff, somebody's gotta tell the computers the decisions to make in a situation with, could be an ethical situation or that has to be programmed. Maybe some of that's becoming generative now.

Well, you know, for whatever reason, I primarily, if not solely, get called upon by non-tech native companies. So the vast majority of my clients are non-tech native Fortune 500 companies. Healthcare, CPG, insurance, the mining industry, surprisingly. Uh, money.

Okay.

E-commerce. But like OpenAI is not calling me and I think it's because they think they've got it handled.

Well, there's a little bit of bro culture and arrogance there as well, potentially where the, uh, yes, I said they think, oh, everything's, I, I put it nicely by saying they think they've got it handled.

Um, yeah, but yes, that's exactly right.

Well, this has been awesome. I appreciate your time and I love the journey from, you know, whole lot of love and in a van and fireworks, Bad Center Lives Get, you know, to academia, back to entrepreneur, like very, very cool. 'Cause the journey's a big piece of it, at least from like, It's one of the most interesting things I learned from thought leaders is how the hell did you get here? 'Cause like you said, you, you know, you asked the heart surgeon how they get there. We know they went to medical school and then they did this and they did that and whatever. But, uh, thank, thanks for sharing the journey and your story. I appreciate it.

Yeah. I'll, I'll say one last thing about how I got, how I got there. Just try a bunch of things out. I mean, you know, I'm not even, it's funny that I'm in a sort of risk management role or something along those lines, 'cause I'm a risk taker. I don't, you know, I'll just try stuff. I'll just try it out. And so part of that journey was let's hire a PR person. Does it— let's see if that works. Let's try email automation. Bad idea, but let's try it. Let's try LinkedIn automation. Let's try this relationships person. Let's— and so that's nothing that academics usually don't do, right? They stay, they stay in their narrow lane. They're like, I know this thing. And then they write a billion articles on that thing. But as, as you know, I know, you know, part of being an entrepreneur is let me just, I'm gonna start a podcast. Let's see how that goes.

And so part of it was that sort of firework, but academics, the ship it, if you will, mindset has been beaten out of them in order to be a successful academic, right? It's ready, aim, aim, aim, aim, aim, aim, and then you make fire. Yeah. Ready, fire, fire, you know, ready, fire, aim. I, so I just think it's just a different capability to develop a different mindset. I mean, academics are perfectionists, which is great.

Downside, right?

Back office is they don't ship enough. They don't break things. They don't, you know, they're afraid of, you know, what happens if you publish an article and you're wrong? Okay, well, nobody—

right, right.

Like, get over it.

You know, it's a, it's a lot, that's a lot of hedging.

The 7 people that read it will probably make stuff about it.

Yeah, exactly. Exactly. I've had to tell, I've had to sort of teach myself kind of like, no one, no one cares about you. No one cares. No one's hanging on every word. No one's reading your words like they, like you read your words. Yeah. Just chill out and give it a go. And if you start a podcast and nobody listens, okay, who cares? No, it's not a big deal. That didn't happen. But you start a Substack, which I did recently, and, and no one reads. Okay, who cares? Like, would you just give it a shot? Doesn't matter.

That's gonna be in your obituary. I've seen they're already writing that. That's gonna, that'll look in history. He, he scares Substack and no, he started a Substack.

Yes, I will be. Yes, I will certainly be renowned for my Substack. There's no question. Yeah.

Yeah. This has been great. Appreciate your time. This has been awesome.

Thank you so much. Yeah, it's my pleasure, Peter. Thanks.

To learn more about Thought Leadership Leverage, please visit our website at thoughtleadershipleverage.com. To reach me directly, feel free to email me at peter@thoughtleadershipleverage.com. And please subscribe to Leveraging Thought Leadership on iTunes or your favorite podcast app to get your weekly episode automatically.