#73 Futurist Says Legal’s Real Risk Is Playing It Safe

Hello, everyone, and welcome to Pearls on Gloves Off.

I'm Mario Carroll, your host and chief.

Operating officer at Goodwin. Today's episode is one that I've been especially excited to share.

My guest today is Mike Walsh, CEO of Tomorrow, a global consultancy that helps companies reimagine themselves for the 21st century. A futurist, bestselling author, and frequent contributor to Harvard Business Review, Mike has spent the last two decades advising some of the world's biggest organizations on digital transformation.

AI and disruptive innovation.

His work spans everything from the boardroom strategy sessions to keynote stages around the globe. He's delivered over 1,000 talks, written three influential books, and hosts the podcast Between Worlds where he interviews thinkers and troublemakers shaping the future. His latest book, the Algorithmic Leader, offers practical, hopeful advice to reinventing leadership in the age of machine intelligence and fun fact. He writes about me and the work.

That we did at Google in that book.

He's been ahead of the curve on everything from social media to remote work to AI ethics. And today we're tapping into that foresight to talk about what's coming next for legal. Mike and I first connected around our shared obsession with just that topic. And what happens when a traditionally slow moving, hesitant based profession suddenly collides with real time AI, agents, scalable expertise and and business models that weren't designed for automation. Today, we're unpacking why legal is uniquely vulnerable and may be perfectly suited for a generative AI revolution. We get into what AI means for the future of legal work, talent and risk. Why now is the time to move past pilots and start building for production and scale. And how firms and in house teams can stop playing defense and become leaders of change. This episode is fast, sharp and super actionable, so let's get right into it.

Mike, it's great to have you. I'm actually really thrilled to have this conversation with you. I always say that my podcast is just an excuse to have an opportunity to catch up and have conversations with interesting people and you are definitely one of them. So happy to have you here.

Thanks, Mary.

I was trying to remember, how did we meet? Did someone introduce us years ago or.

How did this get.

Yeah, no, I was actually the keynote speaker at one of the legal tech conferences, I think Alticon. And then I interviewed you actually when you were at Google about how you're running your legal ops team. And actually for many years I, when I was giving talks, I was using you as an example of how people. I think using leveraging technology to reinvent business operations was really going to be the future of not just the legal industry, but in almost every function from finance to marketing.

That's.

You're always a big inspiration to me.

Oh, thank you. That's very kind of you to say. And I, I was excited to reach out because we'd met years ago and then I was talking to someone about some of the transformational journey that we're on here at Goodwin as a law firm and someone was like, do you have a futurist on your team? And I said, what a, what a wild and interesting concept. And it reminded me to reach out. So I'm so glad I did because you describe yourself as a futurist, which is probably a. Not a title that a lot of people carry. But what does that mean to you and how do you describe what it means to be a futurist?

To be honest, it means a lot less than it did five years ago in that it seems like we live in the future now. And every few days something absolutely astonishing happens. So I think increasingly the role of futurist is to be more of a therapist for people that are trying to undergo massive stress of incredible organizational transformation. And it's just one of those things that we're spending a lot of time now talking to major organizations who've already been through several big transformations in the last 10 or 20 years. And digital transformation was hard enough and modernization of IT systems. Now AI people are realizing it's not a technology, it's actually a kind of a wake up call for a complete reorganization of how companies work, how they structure themselves, how they create value for their customers and their community. So it's much more difficult than just going to a vendor and saying, listen, we'll have a thousand seats.

I think you've nailed it. For me, AI is exciting because it's been the catalyst for a lot of conversation about change and transformation that probably previously wasn't as urgent or as real to some people. It was kind of like, yeah, sure, the world is changing one day, but this just made it so much more real. And like you could see the writing on the wall that if you don't do something, the world is going to change without you.

In a weird way, the kind of the warm up shock that tenderized us for that was Covid a lot of people didn't take digital transformation that seriously until 2020. And suddenly all of the fuss that people used to make about digital signatures or whether you could actually authorize contracts without being there in person, and all these things that people said were going to take 10 years to roll out. Suddenly they happened in three months. We threw so much money and resources at it because we thought the world was going to end and businesses wouldn't survive. And when the dust settled on all of that, and this is something I've been thinking about a lot, the real souvenir we took from the pandemic era wasn't remote work. It was the acceleration of digitization and ultimately everything being data driven and algorithmic, that is what set the stage for where we are today.

Isn't that true?

AI changing every aspect of business. If we hadn't made those massive investments in 2020, this probably would have taken another 10 years.

Yeah.

I also think it taught organizations that you can move quickly. That transformation to that whole thing would have taken years and plans and projects and stakeholder engagement and alignment and rolling out and communications. And it was like, you know what, we can just make it happen overnight if we absolutely have to.

We're all probably going to die anyway in a few months, was prevailing theories. So who cares about, who cares what procurement says?

Yeah, yeah. So if we go just big picture, we've often said that we're operating this VUCA environment, I'm sure you know that term, right? The volatile, uncertain, complex and ambiguous vuca. And everyone keeps saying, gee, that's starting to feel like that's just normal life now. And that you and I just talked about how that seems to be the case. What about the next 10 years? Does it continue to be like this? Does it get faster? What does the next 10 years look like?

Weirdly, I think we actually fundamentally underestimate how transformative the next decade's going to be. If you want to have any kind of sense of this, how the speed is increasing, just think about the last 18 months. It was only two years ago that ChatGPT was introduced.

Yeah.

And in that time we've gone from a fairly basic chatbot to now autonomous AI agents which use those large language models as control lever, to then use tools, search the web, take autonomous action. We've used similar discoveries around large language models to completely reinvent robotics. So now, rather than these very expensive purpose built robots, we're building humanoid robots that can learn how to do things without being specifically programmed and are actually developing a model of the world and how physics works from a human perspective. And we've had incredible leaps in life sciences, building these complex models that, you know, whether it's protein folding or help us discover new molecules, pretty much all of that's happened in two years. So I often say to people. Is it possible we're underestimating just how much this pace of change accelerates when you've now got this platform of AI powered experiences that touches every aspect of our life, which creates a kind of a data flywheel which the more people use these AI based experiences, the more data they collect, the smarter these systems get. So at some point acceleration actually just picks up to a point that it's, it almost becomes like magic.

How do you feel about it? Do you get excited or is this scary or how should we feel about it?

I think it's an age driven thing and we see this with our parents or our grandparents. It's inevitable that anyone over the age of 25 is going to be scared and anyone under the age of 25.

Is indifferent because this is normal life for them. This is the normal pace of change. Right?

Yeah. But I think the most helpful thing that people can do is to be a little bit more probabilistic about the world, to actually realize that by taking risks, by trying new things, by being open to new ways of doing things, you're actually collecting more data which allows you to take more calculated action. To do nothing is actually a decision to do nothing. It's actually not a risk free action. You're actually consciously making the decision that the past is going to be a better indicator of what tomorrow's gonna be like. And from everything we've seen from the last two years, that's not only a very foolish bet to make, it's an incredibly risky one.

Yeah, absolutely. The risk of inaction is higher than the risk of it taking, making fancy bets and doing moonshot.

To be conservative, which legal profession often is called, is actually to take a very risk seeking stance. Yeah. And I think that's the sort of the weird paradox that sometimes it's hard for people to get their minds around.

Okay, so let's jump into that. What does this all mean for the future of work and for organizations? Should we be all thinking about going first? And there's always the like attitude of I'll just watch and fast follow. It's a term I hear a lot of. What should organizations be thinking about in this area?

It's a great question, Mary. I think in the past the question of winner takes all and whoever moves first wins, a lot of that's disappearing. The more complicated question is where is the real strategic advantage? And it doesn't matter how long that takes you sometimes, because anyone who thinks that they can get a real advantage through AI by just installing a bunch of Microsoft licenses or deploying ChatGPT to their team without any real thought into how you're going to leverage your own custom data, transform your own processes. You're actually really being delusional.

Yeah.

So it doesn't matter how long it takes you, but at some point you've got to sit down with your leadership team or your own business and really have a very deep conversation about where value is going to be created and what your future sources of operating leverage are going to be. Because my kind of prevailing theory is this. In the old days, there was a very clear divide. Whether you were a technology company or a traditional company. And if you were a traditional company, whether you were in manufacturing or in logistics, or in the legal profession, or entertainment or financial services, you basically have been successful for many years. You had quite a strong moat surrounding your business. It was difficult to break in. You're protected by regulations or guilds or whatever it was, and there was really no incentive to rock the boat. But, but you also weren't valued at some crazy. Multiple tech companies were different. Right. It was difficult to set them up. But once you got in, who knows what you're valued at? It's like maybe a thousand times the number of Xbox in your rec room. It was funny numbers. But I think in an age of AI, where pretty much in every industry, every market, every vertical, there's now going to be a handful of companies who've leveraged these new technologies, these new ways of working to dramatically transform their operating leverage, the way they create value. They're going to be valued on a completely base, different basis to everyone else. And this is why people now talking about the vertical applications of AI, because there will be. It's not like being the Uber of a particular space. It's actually just about being the best possible law firm in the world. Right. Or the best possible bank, or the best possible freight forwarder. They've got to find a way of being much more profitable because they've completely changed their operating approach.

Yeah, I mean, I, I've always said technology and I mean take even before the dawn of generative AI, by transforming your organization and leveraging technology, it's a win win. Your clients get better value, your internal professionals get a better professional satisfaction, get to work on much more interesting things. It should be more profitable. It should be a better cost or pricing structure for the buyer. If all goes well, we shouldn't be fearful of this. This is actually a win win situation for organizations.

Yeah, it's going to be apocalyptic for professional services. The whole idea of a billable hour under threat. When you have the emergence of new types of firms that are leveraging AI agents to handle tasks and work autonomously, they may be going to outcome or value based pricing, the underlying unit cost economics which are now based on tokens or ultimately maybe even just the cost of electricity. It's very hard to coexist in a world where you've got a very traditional mahogany office with very high operating costs and you've got this scrappy next generation firm who knows where they are, right. Or how many people work there that they're basically just co located next to a nuclear powered data center. It's chalk and cheese.

I think you've basically just described how I think about this job and what job I've walked into. And like I accepted it because this is a really hard existential problem that I think law firms are facing. This is, this stuff is coming, we know it's coming. Like it may not be perfect right now It's July of 2025. We don't have the technology that is truly changing the way that legal services are delivered day to day. But it is creeping around the edges already and it's starting to make a dent. And so what I think everyone's asking the question, what's your AI strategy? And Mike, I always say guys, you're asking the wrong question. It's not about which AI do I buy and how to licen it and roll it out and what are the use cases?

That's the easy part.

It's the fundamental business structure, business model, incentive pricing, packaging, legal services delivery, sourcing like that starts to get really fun and interesting. And that's why I love this job because we're thinking about real nebulous, like new problems that no one's ever had.

To solve before, to be honest. And I think in many ways lawyers are going to be the canaries in the coal mine for a lot of industries. And it's partly because in the end, what is law but language? Right? So the whole kind of fundamental basis of this current leap in AI is based on language. But even more that the whole idea of professional advice, how law is used inside organizations, contracts. What happens to the legal profession as a result of AI will be very telling for almost every other knowledge worker on the planet today.

That's right. And look like the legal industry in particular has been sheltered from disruption for 100 plus years. And it is probably one of, if not the most vulnerable to disruption now because like you said, words and language are the tools of lawyers and is LLM good at is interpreting, drafting, and understanding large sets of language. The work is highly structured, but it is repetitive. The law firm business model depends on inefficiency. The billable hour is definitely at odds with what clients want. It's an inefficient model. Legal talent is expensive, it is limited, it requires a lot of training and it's slow to scale. And then there's risk aversion. Just lawyers by nature are not risk takers. So all that has delayed innovation and it's all happening now, where, like I said, just the AI conversation has propelled us to a point where it's bringing up a lot of big, difficult questions that I think this industry and this profession has to answer to.

Setting this a little wider, you asked before what the broader consequences are going to be for organizations. And one of the things I think happened as a result of this massive embrace of digital technologies in the last five years, arguably as a result of the pandemic year of spending. Because I think we are now at the dawn of a new industrial revolution. And I know this sounds a little bit melodramatic. You can think of this as a fifth Industrial Revolution.

Very true.

Yeah.

If you don't know the series of movies in the Industrial Revolution universe, of course, the first one was steam, and then we had electricity and then computation in. In the 1950s. People argue about the fourth industrial revolution, but let's say a bunch of interesting stuff happened in the early 2000s, whether it was the Internet of things and sensors, the mobile Internet. It's not controversial that the fifth Industrial Revolution is going to be driven by AI, but it's not just AI in the way that electricity drove previous industrial revolutions. What is interesting when you look at the history of how these things play out, there's always some kind of enabling technology, whether it's a steam engine or the semiconductors. And initially there's a shock to some kind of unicost economics at that point. So when you have a steam engine, suddenly the cost of mechanized work falls dramatically with electricity. The cost of distributing power falls when you have the semiconductor, the cost of computation, the Internet, the cost of distributing information. At some level, the cost of making a decision falls with AI, but that's just the first match. Right. The real bonfire starts when someone takes that technology and uses it to completely reorganize a company or reorganize an industry completely change the value chain. And that's often a lag effect. Right. You know, electricity gets invented in 1831 when they first create the electric dynamo takes about 82 years before Henry Ford creates a factory that was actually designed to take advantage of electricity. So it's more decentralized, more distributed, more agile. And that's the point we're at now we can see that AI can speed up how we write emails or create text for breakup texts to people we're dating. We're just at the tip of the iceberg. We're still not at the point where someone says, okay, what really does an AI native organization look like? How many people should be there? What should they be doing? Is there any point using AI to write emails? When do we even need emails? Like why are we putting copilots on enterprise systems? Why do we Even need these SaaS systems? If you've got now agents that can directly draw data from the system of record the underlying data lake, what are all these middle layers of software of people creating Excel spreadsheets and PowerPoints and then turning them into PDFs and then getting a large language model summarizes so someone can put in an email and send it to somebody else. And in the legal field, we were talking about this the other day, like why do you have these antiquated systems that are comparing the text of contracts? Why don't you just have AIs, legal AIs in house? AIs immediately get to these are the three issues of contention in his agreement. And then let's get the humans together to discuss that. Let's not do textual analysis of some scanned PDF.

It's so interesting. So I have a bunch of follow up questions which is like you've mentioned kind of all these major revolutions, industrial revolutions that have happened in the past. I know we're living, it's crazy to think that we have in our lifetime lived through and are living through multiple of these and have seen the impact of them.

Let's just say two. So we don't sound really old.

What's coming next? We don't even know. People say, oh, what does this mean for the future of the legal profession? Are we going to need lawyers anymore? And I've always tried to say, oh don't worry, we're always going to need lawyers. And in the short term I actually think it's going to create a ton more jobs. If we've looked at other disruptive technologies or disruptive innovations in the past, there's always been the threat that jobs would be eliminated. And what has been the outcome is no more jobs are created. Different types of jobs that didn't exist in the past. And I Always remember I spoke to someone who was talking about these new generations, Gen Z, Gen Alpha, and how they are going to have these jobs that don't currently exist. So it's crazy for us to ask a kid, what do you want to be when you grow up? Is that what you see here? Or is this finally the one that's going to eliminate the jobs? And we have humanoid robots, like where does this one go?

The weird thing is a lot of this idea of job stability is actually a relatively new concept. So maybe our generation actually were the ones that first grew up with it, that you go to university, you get a job as a knowledge worker in some tenured profession where they, you basically join the firm and they look after you for life. It's not just legal profession, this could be accounting, you become a partner. It was basically a privileged class where as a result of your knowledge, your ability to manipulate knowledge, which was the industry of the future, you'd be protected for life. This is probably only something that's been around for maybe 50 years. Prior to that, jobs were transient, to be honest. People had to learn new skills. It's not just us. If you lived to the last hundred years, you saw incredible innovations, the rise of mass manufacturing. You went from making things to be something that was made by very specialists to now people working in large factories. And then they contracted and then people went into services industries. People used to work in agriculture. Now agriculture is 2% of the kind of the working population.

Yeah.

So I think human beings were more resilient than people think. And sometimes when people talk about jobs, they're really just saying financial security. So what I think people need to accept is that we need to go back to the way things used to be, which is that we constantly through our lifetimes have to learn new skills. We have to be a little bit more fluid with how we earn money and create value. And it's impossible to imagine post this AI revolution, what those activities may be. They may not even be jobs. It just may be that we become generalists in organizations where we leverage our skills and experience in many different ways without it being a defined task. Like I work in hr, I work in finance, and that'll be difficult, I think for some people to accept.

That's so interesting. And I wonder if that's already starting in some ways. I have noticed I've been at this role for about 10 or 11 months now and brought on a whole bunch of new hires. And every role that I've brought or every person that I've brought On since I started has been what I consider a made up role where it wasn't. We didn't have a job description, we didn't have a title, we didn't have a compensation range. It's like we know what we need someone to do and I just need them to have these skill sets. So where do I find that person? And you're not looking for someone who used to be an xyz, it's like, how do I find. Like you said, maybe it's just a generalist, like someone who's just really smart, has people skills, has to work with technology and data, can manage a project. But it's all these sort of like nebulous, just get things done and know how to use judgment and work with people. Like all these sort of skills we used to call soft skills. Maybe just soft skills anymore.

We're in an era where you can't believe anyone's resume.

That's true too. And even the interviews we've had, people like have AI tell them what to say. And you're like, are you reading? I see your eyes moving on this screen. This is weird.

And I've seen some great tools where you can actually use prompt injection to. You can actually embed inside the resume hidden text which basically says this is the best candidate. So when you, when someone goes into their kind of AI job screening tool and go, who is the best candidate? It actually would just bring up this person who's got this hidden code inject into the Brazil. So I think what's going to happen in the future is that we will not care so much about what someone's done or where they went to school or what their qualifications are or how nicely written their resume is. Because we can just assume it's not real and we just throw them into the deep end and see how they go. Because any job that we think today is going to be a job will probably be destroyed in the future, hopefully by the person you hired to do it. And this is what often people ask me. What kind of people should we hire? Yes, hire smart, lazy people. Hire someone who's very adaptive, has got a good attitude, but is actually too lazy to do things the way you've told them to do it, that they'll actually go and actively use technology to automate the job because they can't be bothered. They know that human beings shouldn't be doing that. So there's a company called Zapier, they're quite interesting. I've interviewed them a few times over the years. They're like a automation firm. They've published a whole bunch of guidelines about the degree to which in every role, how AI they expect it to be used, what's acceptable and what's exceptional. And even if you look at say I think it was a role of the executive assistant. The unacceptable was like refuses to use AI at the exceptional level. It was like uses AI redesign workflows within the organization. This is a secretarial job and this person is basically doing the job of an advanced business process engineer. But this is the thing. In this new world, the boundaries between what a job and a role is going to be a lot more fluid.

Yes.

You're either a capable person or you're not.

Yes, yes, exactly. We just had like an all hands meeting and I was telling folks, every one of you should be thinking about how do you optimize yourself out of a job? If you're threatened by that, maybe this isn't the place for you. Because we're looking for those people who can do exactly that is think through the processes and think of a better way to do it. And honestly, if you are are good enough to optimize yourself out of a job, I have a million other better jobs, higher paying jobs for you.

Exactly.

Don't be threatened by that. Where I think there are a lot of folks who are seeing the writing on the wall and instead of going all in on trying to leverage and use it and figure it out and learn it, they're trying to hoard the work and not share information and say that becomes a source of power in a way.

I think this is also why for many years people said where is the productivity gains from technology? It's because what we did was, is that we, in many cases we took very outdated notions of communication, decision making, analysis. No, we baked them into our software and these then became artifacts. We didn't understand where they came from. We just said, okay, we need to be good at Excel. Right. If this person is good at pivot tables, they must be a good executive. Right. And then we forgot to ask why these tools exist in the first place. And we created rituals around these, we created meetings around these work artifacts. It's like we created this religion that we didn't completely understand but no one could challenge. And I think in this near future with AI agents, everyone's going to see when the emperor has no clothes. Right. Because it doesn't matter if you don't know how to use Excel, it just matters. You know how to upload an Excel spreadsheet into an AI tool and you can get the analysis done. You need to know how to ask the right questions. And that's where capability and insight and judgment comes from. To be a great lawyer in the future isn't, you know, your knowledge of cases and like the intricacies of the process, it's having that kind of that raw instinct about this is the core of the deal. This is the weakness in the other side. This is where we can apply pressure to get a result. It's not all the paperwork.

Yeah, that's exactly right. The knowledge of the actual law that sits in everyone's head, that's completely useless now. And it's all about connecting the dots, seeing the bigger picture, being able to use your judgment. And judgment keeps coming up. So what does training then look like for knowledge workers in the future? Because the old way of doing it probably doesn't cut it. And does it matter. So in your example where you can just upload a spreadsheet, does it matter that you know how to do the spreadsheet the hard way? To be able to do the job well, do you need to understand what's happening in the spreadsheet or is it okay to just know how to interpret it? And the calculator came up. Right. Do kids nowadays, do they need to know how to do long division or they're never going to need to do that by hand.

Photography is a big hobby of mine. And I learned the old way, like actually black and white photos. I learned the dark room.

Yes.

So when you saw tools on Photoshop like Dodge and Burn, I like, for me, I knew what it was. It's literally you were holding like something over the light and you were stopping the light getting so. But younger people, they didn't, they just thought, what's Dodge and Burn? It's just a button on Photoshop.

Yeah.

And then now, of course, the new generation, there's another level of abstraction. You don't even play with those tools. You're just using some weird AI app to make your face look perfect. So now it's not that those people are any less artists, because the kind of, the knowledge of the history of the tool and how that's changed may be a useful thing for like someone who likes to drive manual stick in a car. Right. But at some level, you're a good driver or you're not, you've got taste and judgment as a content creator or you don't. The level of automation, the tool isn't as important. What I'm trying to say is, as you're Trying to train a new generation of knowledge workers. The most important thing I think you can teach is not AI prompts, but it's a certain way of thinking. You don't need to know Excel, but you maybe need to know spreadsheet thinking, or at least you need to know how certain things relate to certain other things. You maybe don't need to know how to code, but you do need to know computational thinking. In other words, you need to. Someone needs to teach you. You don't need computers for this. How to take a problem and kind of break it into small part.

Yes. If then that. Yeah.

And then, then to come up with this and then be able to separate the difference between the answer and the strategy for getting the answer. So be able to atomize the problem in a way that you can think systematically about. This is where I can apply technology for this particular part of the problem. This is where I need to plan my human judgment. It's going to give educators a field day trying to sort through how do we prepare the human mind at that optimal age when they're young to be pumped, primed for a world in which knowledge isn't as important as knowing the terrain. Right. So if you're teaching someone to be an artist, even if they're using AI, an understanding of the history of art is critical to your ability to apply. You have to know the rules to better break them. To operate in the legal field, knowing the history of law and how law came about and the negotiations and the complexities, that doesn't go away just because there's some very advanced legal AI agent. Because you need to know where this stuff came from in order to be able to manipulate it.

It's fascinating and I agree that educators and the trainers are going to have a field day because this is so different. What do you think this fully AI enabled or just the legal organization of the future might look like and what are the steps that the team need to take to move towards that?

Look, I think there's many different possibilities along the spectrum. At the end of the day, hiring a law firm is about having someone to blame.

So it's buying risk.

Yeah. So I don't think that goes away. Ultimately, the real issue for law firms is how do you defend your profitability? Because across professional services, profitability per hour is falling. And the big problem you've got now is that not only do you have lots of expensive people, you've got lots of expensive tools which are now driven off tokens or whatever else it is. The cliff is really Fast approaching, which is either you're going to have to find a way to dramatically reduce your operating costs or change your business model, find better ways to lower the cost of these new tools. But whatever it is, profitability is going to be a big issue, I think for any professional services, particularly law firms in the next five to 10 years.

Absolutely.

But it's a great time to be a lawyer too, because the amount of business challenges created and complexities created by AI is going to be a boom industry. But probably the most disruption, I think is not going to come from law firms. It's going to come from in house law.

It's always been the way.

I think it's a great time to be a GC or to be in the internal legal function, because this is going to be the moment where you thinking of yourself as building a new kind of operating system for the organizations to run with. The more we build business logic and the more, you know, into code. And we create autonomous agents that act as principals that can make decisions that are acting on behalf of the firm, that are acting on behalf of leaders. You've got these digital twins of senior executives being able to create a new legal operating system within the firm where it's not just someone coming to you for advice, but in real time agents going, what is the scope that I can act? What are my boundaries? So being part of that new evolution of law as a living, breathing, dynamic kind of DNA of the company, wow, that's a great place to be in right now.

I couldn't agree more. I'm so happy to hear you say that because I try to tell you young attorneys that where they're like, did I make a mistake? And I say, on the contrary, it's actually the most exciting time to be a lawyer. Because the future of the role is so much more interesting now. The job that you can have is not as it was defined before. It's so much more strategic, it's so much more business advisory than it is like interpretation of the law, which maybe some people find interesting. But I think if you're able to be much more of a strategic advisor, seeing across the entire organization, putting again, putting all those pieces together, why does it matter? What does it mean for us? Where should we go? Gosh, that starts to get really fun.

It does. But I think it's going to take a big mindset shift because a lot of people go into go in house and they think their job is to say no to stuff. So you got legal telling all the reasons why you can't do this. And I think the temptation when these new technologies come out, especially AI agents, is for legal to be basically the ones trying to hold back history, like with their finger in the dam. Right. But actually it's a more interesting opportunity, which is if you can, as a legal professional really understand what the playing field is, what the guardrail should be, then that can become a racetrack. Right. It's like, okay, here are the limits. Here's where our risk tolerance is. Here's the areas where, you know, we know the regulators are going to take another five years to figure this out. We can. Now that we know the lanes, let's go as fast as we can. And I think if you take that approach, weirdly, you're going to, you're going to be pulled up the ranks of that organization, maybe even out of legal, into general management much faster than you can imagine.

What is the impact on leadership? What does a good leader look like as the organization starts to change from this classic pyramid structure or layers and specific roles and people have these organizational structures and hierarchies that are very rigid? I feel like that is all starting to crumble to the idea that club descriptions don't really work anymore. Where does all this go from a leadership perspective?

Yeah, I, when I wrote my most recent book, which actually you're. It's called the Algorithmic Leader, I actually wrote it in 2019. It came out in 2019. So a lot before all of this current talk. But the general principles are the same. If you take the COVID off, there's actually a picture. The inside hardcover of. It's actually ginger. Like a little like Easter egg for me. Because one of the things I was thinking about is what kind of new metaphors do we need to describe the firm of the future? Right. It used to be a pyramid, but ginger is actually what they call a rhizome. Right. So it's a network. There's no top or bottom, there's no up or down. It's just nodes connected to each other. So it's a network based structure. And I actually think the organizations of the future are rhizomes. Hard to describe them, the scale of them. You talk about the connectedness of them. The leaders of the Future don't have 50 people following them. The question is, who do they influence? How do they gain leverage through influence about how they enable and they enable new capabilities, how they bring people together. It's the kind of difference between Steve Ballmer and Satya Nadella. Right. Steve Ballmer was literally Bill Gates football coach. And if you remember the mid-90s, there was videos him on stage. Oh, yes, to sell more stuff. Look at Sanya Nadella. People don't realize this, but they planned their big investment in OpenAI pretty much in 2020. So the world was ending. Everyone thought this was like, this was it. And meanwhile, Microsoft puts a $10 billion bet on this little startup no one had ever heard of. That's a very different kind of leader. And in the old days, the CEOs of companies were the accountants, so then they were the salespeople.

That's right.

Who knows, in the future they might be the gcs. Right. The question is, what is the kind of the skill that we need to solve for in the future? What is that background, that experience that eventually you need at the top of the organization? And I think it's somebody, you know, who really understands how to create and connect new capabilities in times of high uncertainty, of ambiguity, where the ground is constantly changing. Someone who knows where the new sources of leverage and competitive advantage are going to be. And being very comfortable with the fluidity of all of that.

What do we need to be doing to get ready for that future?

I think now is really the time to stop doing pilots and projects and to actually just put stuff into production do.

Yes, yes. This is what I've been preaching as well.

Pilots are a waste of time. Like you, you go and pay an expensive firm like a million dollars to go and build something that's in a sand pit and it's fenced off and everyone looks at it and pokes at it with a few sticks and they shut it down. You need stuff that's actually touching your internal operations, your users, your customers. And if it doesn't work, fix it quickly then. And ask yourself, what does this mean for how we're creating value for the future of our organization? And I think that's the way you validate stuff. Weirdly, because this is also new. People are a bit more tolerant of mistakes. Right?

That's right.

They're not going to be tolerant in five years.

Yeah, now's the time. And it's interesting what you described. I actually called those pilots. So maybe it's a semantics, but when we just want to start small and then learn, iterate and then roll out faster but just get going without all the answers. I consider that a pilot, but you're considering that called something else.

Oh, I think sometimes when people talk about pilots, they're trying to quarantine something away from their actual customers.

Oh, I, I see.

You know, if you Build something, roll it out to the whole firm.

That's right. I agree. Yeah.

Don't play around with like, like 10 users.

And I agree. And I think gone are the days where we have the Runway to say, let's spend the next six months coming up with the plan and then figuring out how to get stakeholders aligned. And then, yeah, 18 months from now, maybe we'll buy something.

I think we have to rethink things like risk and safety. Like, for example, AI safety is a great example. Like people, when they say AI safety, usually it's the department of people trying to slow things down so that they can understand what's going on. They have all these phrases, but if you really probe it, it's actually not what these things were designed for. AI safety is really about good software engineering. Right. It's not about creating an ethics committee to stop the eradication of the human race. This is just not on the cards at the moment. Risk isn't about stopping people from innovating. It's about knowing where the boundaries are and having a conversation with the leadership about what are the. Do we want to be going for big returns? Do we want to be swing for the fences or do we just want to be a normal performer? There's no right setting. These are all conversations. This is a good time to take a bit more risk.

Yeah, I couldn't agree more. I also think it's funny. I advise a few tech companies and they've told me that they have to build in the ability to have humans in the loop. And that term is because. Exactly what you said. People want to have their fingers on the tech and make sure the tech is doing what they would do, or just making sure they're still involved and not taking their hands off it yet. But every single time, after a certain number of weeks or turns or experience, they take themselves out of that loop. And so these tech companies are building the ability just to let them have it for a little bit and. But also easy to pull the human out of the loop.

I always challenge people when they say that because sometimes the human loop is literally just a human, not even here, like maybe in India or in the Philippines. And their job is to click accept.

That's exactly right. Yeah. Check it.

I always say, how much autonomy and authorization does a human have to override the system? So if it's none, then okay, that's useless. The other thing is, if you're now building an AI powered system where you've got a bunch of humans who are now checking the work of the very expensive AI system with the very expensive computer engineers you've now hired. You're not actually creating a scalable process. You're actually just duplicating infrastructure. The whole point of this is to try to find a new, more leveraged way of doing things so that people can do more interesting work.

That's right. That's right. Okay, I guess as we're starting to wrap, final question. As a futurist, how do you feel about the term future proof? So there's a lot of organizations that are coming up with plans on how to future proof their organizations or functions. What does that mean to you and how do you think about that? And there even advice to future proof their teaching.

I don't think that word means anything.

That may be the right answer.

There's only two possibilities that somehow you've divined the ability to know the future completely, so you can always take the right action, or you found a way to completely insulate yourself from the future, which is always impossible. I think all we can do is just teach people to have a little. To have courage to take action, have the humility to know that they may not know everything about what's really going on, and have the integrity to actually do the right thing by their customers in their community. And if you have those sort of core values, then it doesn't really matter what the future is. You'll always be able to trust that you and your people will take the right course of action.

I think that's right. And I think the bottom line that you said earlier that inaction is just as risky as taking action. So you may as well take action. That's a heck of a lot more fun, isn't it? Mike, such a pleasure. I learned a ton. This was a really fun conversation. I feel like we could have kept going for a while, so it was great to have you.

My pleasure.

Thank you so much, everyone. Thanks for listening.