AI Is Rewriting the Rules of Work: Futurist Ian Beacraft Explains Why Jobs are Dead

Hey everyone! Today I'm super excited to be talking to Ian Craft. He's the founder and chief futurist at Signal and Cipher, and he is an absolute thought leader when it comes to the intersection of AI and enterprises that we all work for. He has so many deep, amazing insights. I had the chance to watch a number of his keynotes at South by Southwest recently, and it just never ceases to amaze me how much new thought leadership he brings to the table here. So I'm really excited to pick his brain to understand how much of this new technology is really being limited by us and our own boundaries, and how we can break through that. It should be an amazing conversation.

I've been a big fan of your keynotes. I've been bingeing them in the last couple of days. One of the quotes that I wrote down—I don't remember if it's from South by Southwest this year or last year—but as I was bingeing, one of the quotes I wrote down was, "Poor leadership, adherence to old systems and technology, first, mindsets are a bigger risk than AI to organizations." What is going on out there? And can you kind of dissect that quote?

Yeah, absolutely. So, it to me, when we go through times of change, we need to galvanize behind something. And that happens both productively and disruptively. We tend to find something to create opposition towards. And for a lot of people, they see AI as the main threat because it is the easiest thing to point to and say, "That's a threat to my job." I look at that and I see very clearly it's automating pieces of what I do, and that becomes extractive. It's taking something that basically I provided value through that thing before, and I no longer do. And because of that, I, as an individual, am less valuable to that organization.

Now, if I'm continuing as a leader to just say my goal is to create efficiency and scale within the system that we have today, the typical lever we're going to pull is efficiency, which is code for layoffs. And that is essentially how our system is operated for the last 150 years. And it's been able to grow. We've been able to create prosperity in a number of different ways, but that system's changing now. The era of unending exponential growth in existing paradigms is starting to fray at the seams. And when we think about the future through the lens of the past, what we do is we apply old metrics, old ways of thinking, old processes to new technologies, new ways of working, and new challenges. And those things come together and they don't work. But when leaders are so fixed in how they want to approach these things, they're not thinking about how this is different and how they have to take a different paradigm or a different approach to this new type of challenge and new type of circumstances. And that's what leads to the demise of the organization, not just that employee or that department.

Right. So I want to zoom in on two different phrases you use there that I think are really important. You talked about—shoot, now I'm going to screw it up—but you talked about, you know, growth and the ability for us to have these productivity improvements and be thinking about what we're doing differently. And you also said efficiency, and efficiency has become a really popular word these days. You know, efficiency is something, you know, across the public sector, across the commercial sector. You know, it's a very hot word. To what degree is efficiency the right thing to be looking at right now, or a distraction from what's actually going to help us?

Yeah, I think there's a balance here. There's a recognition that organizations have a duty to their stakeholders and their stockholders, and that means you have to look for efficiencies. And if you're not, that's a dereliction of that obligation.

Understandable. Right. So they should be looking at efficiency and they should be looking at increasing productivity. But to do so with the same fervor we have over the past several decades since the beginning of the digital revolution, I think is absolutely, incredibly shortsighted, because what this does is this doesn't just scale individual efficiency and effectiveness. This changes the fundamental boundaries of what jobs and tasks are. We're really reengineering or completely changing what the atomic unit of work looks like.

So, for example, we take a look at organizations as built from people, which are defined by jobs, very specific slotted roles that are well defined. And if I look at an org chart of any organization, I have these mental shortcuts that I can use to understand who does what, where, and how. All of that's starting to change, though, because AI makes it so that I don't actually have to stick within the boundaries of a specific role and say, "That's all you do." If you're accountant number two in the finance department, or whatever that designation might mean to a specific organization, you have a very specific set of roles, responsibilities, KPIs, and remit that you are responsible for. What happens with AI, though, is it makes it so that the skill sets that might sit adjacent to my existing skill sets or responsibilities are now accessible to me. So it starts to put pressure on these boundaries that we keep people in with their roles. So if you're just a copywriter or just an ad person, if you're staying within those boundaries, now with the access to AI and generative AI and other toolsets, it's also almost an abdication of responsibility to say, "I'm just going to stay in my lane," and all of a sudden we have this chaos that comes with people saying, "I have access to new skill sets, I have access to capabilities," but the system around me has not adapted to really make that possible, fluid, and part of the system where I'm not stepping on other people's toes, I'm not doing things without permission, I'm not doing things without support and an apparatus or feedback loop. So what's happening is we have this new technology that allows all sorts of new behaviors within the organization, right? But the organization has no idea how to pull those behaviors and structures and processes together. Right? It's too rigid to actually take advantage of what this technology could unlock.

Absolutely. So, so what do we—what do we do with that? Like, as leaders, where do we start? How can we start to rethink these systems and processes in a way that's more dynamic, or at least can help us harness these possibilities?

Absolutely. Well, what I would say is the first place you should look is not just about how do I do more with less? You can. I'm not saying don't think about that, but that shouldn't be the primary goal, the value you're trying to get out of AI, because that's a race to the bottom. Like, we're all going to get that benefit. And if that's your focus, then you're playing a Walmart game for, you know, a premium enterprise type of environment that's not going to help anybody. It might give short-term impact. So, a quarter-to-quarter thinking, a Western or American style way of doing business, you'll see immediate impact. And I think that when you take a look at what's happening in the balance sheets at Meta and several other organizations, people see, "Oh, less staff, higher margins, more productivity, that's the way of it." They're missing a lot of what's actually going on behind the scenes. So a lot of these companies that are in the space of building the models and kind of changing the way they work have seen this coming around the bend, and they're already restructuring the way that they operate.

For leaders that are trying to figure out what this means for them, I would say, first of all, you need some sort of experiential learning, like just understanding this stuff theoretically might have worked for the past 25 years because you have a lot of understanding of how the digital paradigm works. This is not new digital, you know, connected networks, all that stuff. We've known that since the 80s and 90s. This is a fundamentally different paradigm, a different way of working, a different way of thinking about growth, a different way of connecting software and systems. I mean, we're literally working with quote-unquote software that now replicates and imitates cognitive processes—a completely different paradigm for people. So having some sort of education or experience that gets you into that headspace where you can start to grapple with what those changes are is absolutely necessary. Just reading articles and doing a couple of things on ChatGPT is not going to be enough, because if you're going to competently lead an AI transformation, you, as a leader, also need to have spent that time immersed in that space. I'm not saying hundreds of hours, but at least dozens in that space to understand it, with the proper guidance, what it's going to do for your team, your business. How does it help you answer the questions about what kind of value we're providing as a company? How do we structure our teams? How do we grow effectively in this market where everyone else is starting to go into these adjacent spaces as well? So there's a lot of new questions that I have to answer from that.

Yeah. Well, when you talk about that experiential learning, Ian, do you have a sense of like what leaders are doing? Like, what does that mean? Like, what tools are they using? What does that look like?

Yeah. So the most effective programs that we've seen are ones that are built to be used with the same types of tools they are already using in their environment. So everyone uses Microsoft Teams, Slack, has access to ChatGPT or something like that. 90% of what you need to do can be done with what's in those environments, but just using the vanilla version of it is not going to cut it. Being able to learn with these tools by also building the documentation, the vision, the information you need to move forward is really where we've seen the most value. So to give you an example of a module that we run, we'll work with leaders in an environment where they're working with the AI to build their vision for what AI looks like in the organization, to create a maturity assessment. So, where do we stand, and where's the alignment amongst the CXOs? And it's not just about the education on AI. It's about alignment. And there's a strong difference between agreement and alignment. What happens oftentimes at a leadership level is we agree AI is important. We agree that we're all implementing it, but they're not aligned as to how that happens or where they even are on a maturity index. So having them come together to do that together while using the tools to facilitate it brings a couple of those objectives together, and all of a sudden they start to see, "Okay, here's how the tools can facilitate this work. It doesn't need to take six weeks. It can literally take an afternoon." So you've taken something that might have been a six-month consulting engagement and said, "We're walking out in 2.5 hours with a much clearer understanding of what we're doing, how we're doing it, who's responsible, and what that roadmap looks like." So that's one of the big paradigm shifts we're seeing. So it's more—if I understand you correctly—there's more value in like one condensed facilitated session of asking the right questions, having the right people in the room, than like a protracted engagement on like, "Here's a bunch of recommendations of potential use cases and what you could be doing." Is that fair?

I think the potential use case model is kind of outdated at this point. So I do think that the idea of the thousand-page deck and the long consulting engagements needs to change. I won't proclaim that consulting companies are dead on arrival. I think that's a little hyperbole there. They're like advertising agencies; like, they survive this kind of stuff. The consulting company of the future looks wildly different than it does today. But one of the things that they're often kind of admonished for is this idea of a thousand-page deck. And I do think that the idea of learning being separate from doing—so, having a thousand-page deck, a bunch of seminars, and then finally being responsible to do it on your own—is outdated. We now have the tools, we have the apparatus to learn and do at the same time, while also building some of the most essential infrastructure, as well as documentation and strategy amongst executives. So when you walk out of a session, you should have a clear vision. You should have an understanding of how this impacts you, what your maturity assessment looks like, and what your roadmap looks like. So what we've come to see is that we can take things that should take six weeks or even three months and condense that into one afternoon or a full-day session.

Right. Which is—which is really exciting. And, you know, I think helps us get a lot more like just accelerate our time to value, our time to results. And there's a phrase I want to talk about that you've said quite a lot, and I want to put some parameters around it, which is the tools, right? Talking about AI and the tools. Because, you know, in my mind, when we talk about AI and when we talk about the tools, it's everything from, you know, just go to ChatGPT or Google and write in your question to, you know, this world I'm finding we live in increasingly where every software vendor and their brother is promising you that, "Oh, there's AI in this now. It's an AI PC." That like any crevice we can hide AI in, we're telling you there's AI in it. What tools do people need to be thinking about? How should we be bringing tool-wise AI into our organizations?

Yeah, we're at the very beginning of the development of a very robust ecosystem. So most people are seeing things like ChatGPT, Claude, Copilot, etc., and that's kind of when people say "tools," they think of that, and that's fine. But what we're seeing at the enterprise level is a weaving of that into the basic infrastructure across the board. So for a lot of people, they're pulling in Copilot; others they're pulling in, you know, OpenAI's API into the work that they do. And stage one is just to get people exposed to tools. So that's access to the chatbots. It's a one-to-one relationship. You put in an input, you get an output that is barely silly. Even alpha products for an enterprise at this point in time, like you're just getting your socks on before you put your shoes on to get out the door. And what we're seeing with organizations that are more successful is their leading with use cases that everyone can understand, and then they're building that into the infrastructure of their organization, not just saying, "Can you go learn how to use ChatGPT?" That's basic and necessary, because people can't start to identify what the use cases are going to be unless they have experienced the technology. And I'm a big fan of pushing that down and out. You can have people at the edges coming up with the use cases, not just the people at the top, because the people who are dealing with the challenges, who are the ones actually doing the work, have a much more nuanced understanding of how it should be done and what success looks like for those types of tasks and activities. To get there has to go way beyond just access to ChatGPT. What we found for so many organizations is they'll often say, "Hey, we invested half a million dollars in ChatGPT licenses, and people loved it at the beginning. And then like, nobody uses it. It goes like this." It creates, and then it crashes, as far as usage. And the big part of it is you just gave them a new tool. You gave them barely any training, barely any context. And you said, amongst all the things you're doing, you're overstretched, under-resourced, your expectations are just getting higher. Now you have to go learn a new way of doing things. There's no surprise people are not using the tools, right? So it has to be very clear from the beginning how this applies to them, why it's relevant, why it's going to change their life, not the organization. Like, really put it into terms they can understand and grapple with. And then over time, as the knowledge starts to diffuse across the organization, the infrastructure also becomes more robust. It's almost the opposite in many ways of previous transformation. So let's take a SAP ERP implementation. That's a three-to-five-year process. IT command and control. We install it. Everyone else has to adapt to it. What's happening with AI is we're kind of reversing that in a way. Yes, it is provisioning, licensing. But this is not just an IT issue anymore. This is an HR issue. This is a strategy issue. This is a finance issue. And if you don't have your CEO, CFO, your CRO, and your CEO all in lockstep on how this is being distributed, you're not going to come up with an effective way of distributing the technology, the knowledge, and the application across your organization so that that changes the dynamic of the tool enormously. And over time, it then becomes part of the fabric of the organization, just like we have with all of our other productivity tools.

So I'm—lots to process there. One of the—one of the things you've said is we talked a little bit about it already, but this—there's a fear of people losing their jobs and, like, disintermediated, like, jobs from tasks and augmenting what we're doing. You had said previously, "We're not going to lose jobs; we're going to lose job descriptions." Well, what does that mean? And what does that look like with these AI and with these modern tools and approaches?

Absolutely. So I said that probably two or three years ago in one of my South by Southwest speeches. I said, "We won't lose our jobs; we'll lose our job descriptions." And I've—I've had several people say, "Well, that aged horribly." And I'm like, "Actually, yes and no." Like, I will willingly say there are jobs that are going to be lost. There's no question about that. But the image that conjures for a lot of people is like, "Well, you're going to be on a breadline for the rest of your life. You'll never be allowed to work again because you're completely irrelevant." That's not how this works. There will be some fracturing of the system that we work within, and that's not going to be easy. And it's going to cause some pain for some organizations and a lot of people. But what does happen is it also changes the nature of the jobs we hold. So what I see happening, and this is related to the concept of creative journalists that I've been talking about for a while now, and I can define that as well. But a creative journalist is essentially—I'll pull this back because it's related to the waves of education and we build our careers. So we grew up in a system that said, "Go to school, get a job, build expertise within that job that becomes defensible. And that's what gets you promoted over time. Over time, you start to manage the functions that you're an expert in. You become management and leadership, and up you go." That vertical way of working has been the way we've promoted people and told people to go after their careers for decades. What AI does, as I mentioned earlier, is it essentially abstracts the years and decades of expertise, influence, opportunity, exposure that you need to build expertise in a specific subject. And it allows you to perform proficiently in skill sets that are adjacent to your own, or even some that you never had access to before. I said proficiently; I did not say in any means an elite level. But what we often have to confront in our organizations is, in many cases, good enough is good enough. You don't need somebody with 25 years of experience to do junior-level work. And if I am an outsider adjacent to that role and I can get that junior-level work done, why do I have to wait? Or rely on specialized expertise and resources to get that work done? So that changes this dynamic and the nature of how I relate to my peers, their roles, my roles, and it expands the capacity and responsibility in the remit of individual roles. We're moving away from role-based relationships to jobs to skill-based and task-based relationships to jobs. And that's where I feel like the idea of jobs are not going away. I even say in that same speech, "Jobs are dead. Long live work." And one thing that I think we're so stuck on and say we need good jobs, we need—you'll hear every politician say, "We need to bring good jobs back to America." And it's not the jobs; it's the work that we need. If we're so focused on jobs or already narrowly defining ourselves and oftentimes attaching ourselves to things that are not coming back, like we're—we're not really going to be bringing back the coal sector that way. Everybody is talking about it in many ways. That's kind of a train that's moved along no matter what we do, and other jobs are the same way. But we talk about the work as it relates to the tasks and roles and things like that that need to be put together for the future of work. That's where we can actually make some traction. And that's why I do believe, like, we're not getting rid of jobs. We're getting rid of the definition of the artificial boundaries that keep you in a specific space in your organization. And that's how I see organizations evolving significantly over the next decade.

So if that—if that is the case, there's like tons of wild implications. And I, by the way, I think that probably is the case. But there's like tons of wild implications from the education system to our careers to organizations. But in terms of impacts in—and maybe feel free to tell me this entire paradigm is wrong—but where is there more risk? Is there more risk for, you know, junior employees right now who, you know, their general skill set and their baseline level of knowledge can be replaced by AI? Or is there more risk for people who are 20 years into their career and they have a deep skill set that maybe now is almost commoditized because you can ask AI how to—you know, I'll pick on data scientists, you know, just for example, because these are historically high-paid roles that have a lot of schooling behind them. And if you can start to get some of these answers, some of this validation information commoditized by AI, what does that mean for people in these careers? So where is the risk greater, or is the entire paradigm just misplaced?

It's so—they're both under an enormous amount of strain right now. The one that's most present that we're seeing happen with more frequency is those who are at the beginning of their careers are at highest risk to this exposure, because you can replace a lot of the things that they would do. Now we're kind of—there was this unspoken contract that when you leave school, you would continue your training almost like a vocation in whatever place you would go in and learn. And there's an enormous amount of teaching, mentoring that goes on with that. And when I can just consult ChatGPT and get it done and not have to mentor it, I'm going to do that. There's just no question that that's going to happen in my case. So junior roles are already starting to disappear. The capacity and capability that juniors have are starting to disappear. And the challenge that just adding the technology to and giving juniors access to that doesn't help a lot because they don't have the experience and the frameworks to understand, you know, when I'm doing some exploratory research on this, what matters, what are the things I should be looking and honing in on? And they just—they get overwhelmed by the sheer volume of things they should be looking at without being able to find the right signals to hone in on.

Now, the other part that's true is that middle management is also getting hit really hard with this stuff. Because if your role is focused more on creating alignment, checking in on organizations, or checking in on your employees, doing a little bit of mentoring here and there, but more so the things that are around productivity and efficiency of a team—so not the leadership level, not the visioning—right? But just like the operations of the company that is directly in the line of fire of AI, and what that changes is it—the stuff of being a boss is also starting to go away. What that makes space for is for people to step into actual roles of leadership. So we're seeing this layer of middle management that is directly in the line of fire, but also collapses the organization a bit, where the line between junior or more junior people and leadership is also starting to become thinner and thinner and thinner. But it means

More direct contact with people who can spend time in the space of being leaders versus bosses. Can you, can you just unpack that a little bit for me? What is it? I'm I'm actually a little bit surprised to hear that that the some of the team management piece is something that I can do. What what are the tasks that you see as being, you know, no, now doable to AI? What does that look like?

Absolutely. So a lot of what happens at the management level is facilitation and alignment, and things like facilitation are easily done by AI. When we actually work within Signal and Cipher, the projects that we're working on are known not only to us, but also our AI and our project management system. So I don't have to check in with my co-founder, you're like, where are you on this? It knows. Therefore, I know, so that the amount of meetings that I'm having around alignments and approval have dropped like 70%. So the wow, these big parts where we call it corporate waste, these items where you're waiting on specialized resources, they come available to do the work or to have a moment of someone's time to say, do you approve of this? Or having those CYA moments of does legal approve of this? Those are actually starting to disappear as well because the systems can check on these things.

Now, none of these things are bulletproof or faultless yet. So in the beginning, they actually create more work in an apparatus. So there's more work for bosses and middle management to work on. There's more work for implementation. And and we see that almost in every transformation, it actually productivity dips and effort and investment go up during that transitory phase. But when you get past that and start to see some of those benefits, it changes dramatically how you operate. So we'll go into organizations. We'll see, you know, 25 people sitting at a table for an hour-long meeting. And the first thing I think of is like, three of you need to be here, and this is so expensive, like you're wasting so much more. And for us, it's gotten to a point where we say that the small team is the ultimate flacks. It means that if you've got a small team that can really run effectively and you're using your tools and your infrastructure appropriately, then you can move so fast, you can make decisions confidently without having to consult everybody. And most of those meetings are really about CYA. It's about, can I distribute the liability of this decision across a group of people? So it's not just my fault, right.

So so with that in mind, you've talked before about this notion of augmented teams and being able to use some of this technology to get more out of an existing team or even restructure an existing team. What does that look like in practice?

Yeah. So there's a couple of things that we lean into to to help augment teams. The first level is just learning to use the tools. Like just by doing that, you're already moving 10 to 20% faster, more effectively, better thought partnership with AI. But the next level is starting to actually encode your own knowledge. So that can happen at a team level or an individual level. What I mean by including it is taking things that you've written, whether that's briefs, emails, contents, and starting to turn that into something that the large language model can work with and understand. It's a bit like training a Llama on your own assets, and that becomes a bit of a digital twin. And we do this at the team level, the individual level and the organizational level. What most people aren't seeing right now, we're seeing a lot of that happen at the organizational level. But when we augment an individual and say, okay, I've taken everything you've written at it, not everything, but the the highest signal, highest quality stuff that you've written, content about who you are, what you've done, etc. and turn that into a document that sits on top of the large language model, and that becomes the filter through which you prompt. It's to fill out a filter and going in and going out. So it's going to augment what you're actually prompting against. It's also going to filter the responses that the LM gives you. So everything you do is going to be in your style, in your tone of voice, with your strategic understanding of the business. And that can really expand your capabilities in a number of different directions. The same thing happens with the organization, which helps enormously with things like brain drain or bringing people up to speed. One of our goals is as an organization, if I bring someone in off the street, I want to be able to get them to the same level as everybody else within less than a month, and on day one, they should be able to write an email, the tone of voice of the company, that should be able to manage our social media presence, all that kind of stuff, because we've built a layer on top of the large language model that already has all of that encoded, rather than having to teach someone to do that from day one. So it's doing two things. It's augmenting the individual's capability and capacity. And it's also removing all the friction of having to become up to speed on how your organization works. Right?

So I mean, I it's such an interesting approach, and I can, you know, immediately see like huge transformational organizational benefits, you know, from an efficiency and effectiveness perspective by doing something like that. Are you finding that there's resistance at an individual level to that? Like my concern would be that people say like, oh, you're trying to like download my brain into AI and then get rid of me. Like, I can imagine a world where there's angst. Is that the case? And if so, how do you overcome that?

That's the very first response most people give is like, hey, this this is my value. Like, this is why you hired me to do this. And that becomes a conversation between the organization, the individual, and a contract of this is yours, not ours. So the goal with the data is for the organization and a team level data that is owned by the organization. The individual level data must stay with the individual. And this is a this is a personal philosophy of mine. I honestly don't believe that we can move towards a future paradigm where this is a part of the way we do our work. If that agreement doesn't stand a place. So it's like when you move from organization to organization, you take your experience with you, but you're not taking the files you worked on at at the office. If I'm going to encode you and your thoughts, it's just like saying, as an actor or a voice actor, I've encoded your voice and no longer have to give you credit for what I extracted from you. That, by definition, creates an antagonistic relationship between the organization and the individual. I firmly believe we should own our data, so that's what we have encouraged and facilitated. But yes, that is typically the way people look at it when they first start. It's like, okay, the company's gonna own everything about me. And right, sometimes the company are like, oh yeah, that's that's right. We could really do a lot more with a lot fewer people. Like, no, that that will absolutely one destroy morale and to no one a lot work with you. Right.

So are we. It's such an interesting idea. And I and I really. I hadn't heard that before, to be honest. And I I talk to people in this space all day, every day. So the idea of having this, like, this individual, you know, digital twin or AI-ified, you know, likeness or, you know, data model of you right now, I have to imagine if most organizations propose that to a staff member, that'll be like the first they've heard of it and they're like, whoa, you know, what is this? Are we heading toward a world in the next few years where this is going to become the norm and, you know, everyone will start to be literate around this and, you know, expecting this conversation almost?

I don't know, and the reason I say that is not because the technology's not ready for it, not because it's not possible. It's that the the limitations of how our organizations grow are not just technological. There's so many different other constraints as to why organizations change the way they do and why technological adoption is so slow. There's a concept that I love called martech law, and it's it's about the difference between technology moving exponentially and people and organizations develop logarithmically. And what this does is creates this ever-increasing gap between what is possible with the technology and what the companies and individuals are actually capable of. So we're looking at potential versus practical reality. And the thing that's pushing this curve so much lower than this is infrastructure, technology, culture, decision debt, technological debt. There are all these constraints in an organization that dictate how high that logarithmic curve can go, and how far you can push that up so the technology can move as fast as you possibly could imagine. We are not going to be able to integrate and adapt it as fast as it changes. We're already seeing that right now. We're right. You'll see some people who are just, you know, they've 100x themselves with what they can do, whether possible, whether it capable of doing and then everyone else is looking at them like they're an alien. And that's because they, as an individual, have leaned into it and are already adapted. A lot of the stuff, they've already had the experience to help them do that. Usually, you know, really good engineers and developers can lean in and the way to do that. But if you're an account person who doesn't have that expertise, you look at that and say, there's all these limitations preventing me from doing that. Organizations work the same way. Some organizations like ours are small. We're built for this where native AI, whereas a lot of organizations we're working with that are Fortune 500, don't have any of those things, any of those qualities that will allow them to say, we're going to be AI native tomorrow. Right?

So, you know, this brings me to another central question that I was excited to ask you about, Ian, which is who do you see as being the winners and losers of this disruption? And I'm deliberately asking that in kind of the broadest possible sense.

Yeah, it's an interesting question for which I'm still forming an answer myself, because the initial thinking is like, okay, if we don't need big teams, then we obviously don't need organizations that are 200, 300, and 400,000 people. And, right, all these startups are going to come and take their lunch. And it's a lot more complicated than that. There are other structures besides just size keeping the the current winners entrenched in their space. So let's take like a chemical manufacturer for for instance, like there's a lot of the corporate work that can be taken away by AI and made more efficient. And you can use machines for that. But there's physical apparatus, there's mechanical apparatus, all that needs to be done, this distribution, there's geopolitical elements to how these companies grow. And again, that gets back to the practical limitations that shape the growth and change of these organizations in new paradigms. So I don't think it's just as simple as, okay, you need smaller teams, fewer people, companies will shrink and startups will come in and eat lunch. Of those who don't move fast enough, speed is one variable. It's a very important variable, but it's just one. What I do think, though, is companies will get smaller. But I also think there will be more startups and more businesses formed than ever before. If we just look at the trajectory of the statistics, even since Covid, we've had a massive increase in the number as corps and LLCs formed more than any time in history, and that's likely to get even easier as time goes on, because the ability to form a company again gets easier with AI, the ability to form a team gets easier with AI. So I think freelancing is going to explode even more than our. Yes. So an acceleration in the existing trend, the ability to open businesses, the things that keep people away from opening businesses is going to almost evaporate. And I think that the opportunity to start creating these entities for even short time periods of times for more specialized use cases is going to become a thing, too. So I could easily see the number of businesses built in the next ten years 100x-ing, not just multiplying on that, because we're also using agents for that too. We're not using agents for employees, but we're using agents to build the business. So if you think about how you can scale that, that's I'm still wrapping my head around what that looks like. As far as what is the economy look like, what is and, you know, how do we align that in terms of geopolitics and how that becomes the way organizations shape as well? It's a lot of things that, you know, still not shape themselves yet. So I don't think it's just a small organization's more business. But that's the closest thing I could find so far. Right.

And I'm glad you had that level of clarity because it's it's easy to just end up in the mindset of, you know, smaller organizations eating your lunch. You're done, you know, good luck. The when I think about the implications of what I think we broadly agree upon, which is it's going to be way easier to start a business. There's already more businesses happening. That's really good for consumers, I would imagine. And it does mean more competition for incumbents. And I like I really like your point about, well, it's not just they're going to eat your lunch because there's more to it than, you know, just just the speed or the efficiency. They're you've talked before about the need for transformational change versus just strictly optimizing what incumbents are doing, how transformational do we need to be thinking, and what's the best way to get into that mindset? Is it creating like an innovation incubator in your organization? Is it trying to start your own, you know, kind of funded startups? Are there any kind of tactics you recommend with organization?

Absolutely. So I think I would encourage organizations to be radical with their thinking and practical with their approach. So there's there are too many people. So you kind of need to break burning all the ground, start fresh. There's no enterprise that says we're profitable. We're doing just fine. We want to disrupt that. Nobody says that. But what I do think is, unless you are radical with your thinking, you will not be ready for the disruptions that are going to come. So these technological transformations that happen at GPT level. So general-purpose technology start at the infrastructure level. So we've seen disruption with technology and the technology that we use. So electricity did the same thing. And OpenAI did the same thing with GPT. So we know now we're all using it. But over time those disruptions move up a level from infrastructure to application to industry. So if you're not okay I guess I guess it is explosive. But if you're not thinking radically about the transformation that can happen at each one of those levels, and also the transformation that can happen to your industry, and you're just focused on the data, what you have now, you're missing one of the critical shifts of transformation in the business. And there's a theme that's becoming more popular right now is going moving from insight to foresight. And when everything is changing around you, insights valuable, it's how you create structure around a business that you can take to market. Foresight is about how you avoid getting disrupted. If we're not looking forward and we're still letting yesterday's mental models collide with tomorrow's technologies, that is how we lose. But if we are radical with the way we think, with the IT ability to test different business models, put things to market faster when we might not previously get that data and that feedback loop as fast as possible, we're going to learn more about that unexplored terrain way faster. So I wouldn't say go and disrupt your your $1 billion, you know, revenue line, but you absolutely should be incubating things that will because there are hundreds and eventually thousands of other startups that are doing exactly that. And you have no defense against that if you're not thinking in that way. So think radically, approach practically. So that next step goes, okay. So what do we do to implement this? Is it Tiger teams? Is it small skunkworks? All of those are viable. I do believe that having in its transformation, you need to find people who are leaning in and are self-selecting as the people who are like, I'm all about this, I want to do this. Don't try and convince a bunch of people who might not be invested in the nest to be the first ones through the door. They will be unenthusiastic about it. They don't have the willpower to get through the challenges. It's going to be hard, and they're going to fail a million times before they get it right. If they're not already passionate about this, they're going to stop at the first sign of trouble. Those people can be followers of the people who lead the way. It's not that they're irrelevant. You need to find the people who are like, I want to be the person who kicks the door down. I want the first person in the room, and those are the ones you want to build your teams around to to think about these things and build different ideas and find the tinkerers. Find the people who may not be the developers or the engineers who are already tinkering with the stuff. There are so many people who are using AI and building their own agents or creating, you know, side businesses on the weekends who could also be resources for this. And that's the culture that will create new opportunities, new business models. And they're going to learn what these new paradigms will look like by doing the work in that space. That then can be diffused across the organization. And that's the second most important part. Once you have the knowledge, do you have the infrastructure set up to diffuse that knowledge as fast as possible and as thoroughly as possible across the organization? Otherwise, it just is. Compartmentalize it. Compartmentalize it. It dies on the vine. Right.

And I'm glad I'm glad Ian, you used the word culture. Because I'm curious. We talked about martech law. We talked about the need to I don't know if we use these words, but bend the curve upward to try to keep pace with technology, to try to compete, you know, to what degree does culture play a role there? Is it the most important thing is that in the top like and if it's not the most important, what is the most important?

I do think it's the most important thing because if you if you don't have a culture or can't create a culture that is willing to lean in and say, hey, things are going to look so different in the next couple of years that we won't even recognize it. It's up to us to make that change. You're not going to get there if everyone is waiting for the vision to be given to them to take action, it's already too late. And that's one of the biggest challenges is a lot of organizations. We built this expectation that when the CEO gives the vision, then people act and people stand there. If people aren't leaning in and saying, I'm in R&D too, like I I'm actively in research and development of what my own role looks like in my organization. My own profession looks like because you're going to encounter this no matter what role you have or what company you work at, you go work somewhere else. It's still going to find you. So we as individuals have to take ownership over this if we want to maintain relevance in this space. This is not an us versus them up versus down organization versus an individual issue. It's a collective one, right. So if that's recognized in a healthy way within an organization, that creates a camaraderie, a collectivism that can move an organization forward, right? If everybody's kind of sorry, I'm going to use the silly like, you know, the silly idiom about rowing in the same direction, but having that purpose and everybody kind of banding together to move the organization. But it's so true and so so spot on.

Yeah. I want to, you know, with that, I want to come back to, you know, another kind of quote you had which is moving from insight to foresight, which I which I love, by the way, where does foresight come from and can it come from? I because my sense is it's like a lot of these tools, they can summarize what's already known. Right? Like they aren't necessarily taking you forward, or they're just telling you the sum of what we know up into this point where where does foresight come from?

Right. So I would actually disagree with that a little bit. Okay. The the my perspective is that large language models are commodifying like they if you just use the large language models, there's a period of probably two more years where you can have an advantage over many of your your colleagues, but over time, it's just gonna be like, I use email, think big whoop. So does everybody else. It's literally a commodity. And the difference is going to be how do you use it? And then how have you encoded your knowledge into doing that. But specifically on the foresight piece, it's it's about searching for signals and how you combine things as the user. This is a place where we're still very much in control. I don't think this is the kind of thing you want to automate. What you can't automate is searching for signals of change. So foresight is really about finding those those data points that are outside of the normal distribution that say, this is different. Like you should you should pay attention over here and validate whether or not this is something that you should be investing your time in or concerned about and in foresight. If you talk about formal foresight, there's probable, plausible and possible futures. There's a whole bunch of structure around it to create really good thinking about what's to come. But I think that can overwhelm and overcomplicate people. We all have a responsibility to think about foresight. What does my role look like in a world where I actually don't have to go to six meetings a day? Oh my gosh, sounds amazing. But what's my new responsibility? Because it's more on me now, right? And like not thinking that through puts you on your back foot and it makes you subservient to the vision of whatever else is happening around you. So when it comes to foresight, we should be thinking about, well, what if this happened? How would I react? And it's not about fortune telling or predicting the future. It's about seeing the signals and patterns that are starting to arrive and understanding the scenarios of how might that affect me? How might I react so that when something does actually come your way, you can say, oh, you know, I've seen something that looks like this before, or this rhymes with something else we've already thought through, and you're adapting versus reacting. You're proactive versus reactive. And I think the best organizations in the world do an enormous amount of that. The ones that don't are the ones that really do get caught by surprise. And we see a lot of enterprises in that space right now. But I do think that the the foresight is where we need to lean, because it's also where we can have a lot more of our human agency using the models and the AI and the tools to bring the data into us, to help us identify what those what our what's different and say, what does it actually mean to me? And using as a thought partner in getting to getting to clarity. Right.

So, you know, on the note of what it means to me and to to be honest, you know, I'm I'm surprised and very intrigued to hear you say that more of the foresight can be done by these tools than maybe we imagined before. And I'm curious, and this is sort of a self-serving question for for both of us. But what role does an organization like Signal and Cipher play in this world? Right. Is this something that all other tools can do it. I can do it. You know, we don't need partners to help us with this. Where does an organization like yours come into play to help actually accelerate, you know, traditional enterprises?

Absolutely. So the to create a little more clarity around that. I don't believe that the AI tools can do them do this on their own. But I think that they can facilitate our own work in coming to an understanding of what possible and plausible futures could look like. So a lot of the research that one would do to do, you know, future scanning and signal scanning to find these opportunities we should be looking into can be massively accelerated, scaled and assisted using large language models. It's still up to us to say, how does this fit with my strategy? How does this fit with the market dynamics that I'm

Seeing play out? So it expands and augments our capability to do it, and it makes it so people who are unfamiliar with this can dive in even faster. So it's less intimidating to get into that space.

There will be more and more automation of that as time goes on. And who knows, maybe in five, seven years we could actually say, okay, just go run and create, you know, do a Monte Carlo simulation times a thousand for me and pull this all together and then give me that data and tell me what my future looks like. I don't think it'll be that simple, but there could be a world that looks like that. But for the work that we're doing, our focus is on helping organizations get that to become part of their culture. So it comes from training, that comes from building that data layer that goes on top. A large language model encoding their knowledge so that they can understand how the signals that come in from the outside world are going to impact them. How might they respond to that? And also scaling the internal workings of the organization so they can be more efficient and effective? Things we talked about at the very beginning; we're not eliminating that, but what we're doing is expanding the capacity and ability for them to operate in spaces they never could have before.

So where that changes is organizations might say, well, okay, I can use this. And now I only need a three-person marketing team instead of a 25-person marketing team. Company A might do that. Company B might say, hey, for the last three years we probably had, if we go back and look at our backlog, 300 or 400 products or projects that we would love to pursue, tested and gotten data on, that there's no way we'd have a team of 500 that we would need. But you know what? Now we could do that with 25, and all of a sudden running simulations, right? Putting products together, testing things, getting that data becomes possible at an enterprise scale from a small team. And you're exploring new opportunities in unknown territories. So you have this expansive mindset versus a contracting mindset. One works very well with an industrialist capitalist mindset. One works very well in a time of transition where new markets, economies, and form factors are starting to develop. But we don't know what they look like yet.

So that's what we encourage companies to do, is say, rather than saying, you're going to lay off 50% of your workforce and do more with less, do more with the same, expansive skill sets, right, expand capacity, expand possibility. And that is really where we see the most value, and I want to talk—I love the—I love the dual approach there. And it makes—it makes complete sense to me.

I'm curious with the—the second scenario you talked about where it's 25 people or 500 new products or tests or what have you. One of the things I've found, and I'm curious if you've seen it too, or you've seen something different, is in this emerging world where technology isn't the limiting factor anymore. And it's like, if you can dream it, you can build it, you can test it at some point, the—like, the bottleneck becomes the market, or it becomes your staff, or in some sense, it's people's ability to—to like, actually try and digest new things. And my sense is you have to still get back to like prioritization in some capacity. Because even if the technology can give you 500 new things, you're going to be limited somewhere. Do you buy that, and what are the implications?

I absolutely do. I think that what that does is it shows another weakness in the current paradigm for the future that we're trying to create. We go through these phases of oftentimes 150, 200, 300 years, where the economic paradigm also shifts. So the one that we're currently in is identical to the one where we built the steam engine and connected geographically disparate places; the metrics that we use are still the same ones that we used, with some modifications, when the steam engine was a cutting-edge technology. So what that does, it shows that the paradigm that we're in is kind of the ultimate bias. A paradigm shows you what's important; what do you measure, what questions are worth asking. And all of those are still very much directed towards the capitalist system that we have now. And I'm not saying this to start a capitalism versus socialism versus Marxism type of argument. It's what is capitalism 8.0 look like in order to start to expand its environment. So these new types of businesses could become possible. So there will be fully autonomous organizations that have zero humans involved. And what does that look like? It's not a full replacement for humans. That would be like saying the digital office replaced paper. It obviously did not. That digital killed analog; analog is still—is absolutely decreasing, but it's not zero. And it won't ever be zero, in my opinion. But it creates this fragmentation of what we saw as like the dominant paradigm. And it creates space for coexistence of all these new middle models, mental models, operational models, and economic models. We don't know what those look like just yet because we haven't seen many of them succeed. We're seeing some signals when we look at companies that are, let's say, on the lean AI leaderboard, which is a reference I love, you know, the average 3.3 million average revenue per employee. You know, time to scale is absolutely insane. And we look at these organizations that are AI native, they're starting to show what some of those paradigms could look like. If you extrapolate that from, you know, ten people, 50 million AR, and say, what would it look like with one person, 150 million AR? What apparatus and infrastructure would you need? What would it look like to operate as that individual? And that can give you some of those signals I was talking about earlier of what possible, plausible futures could look like. But I do think that how we look at capitalism today is going to change dramatically. And it's not just a technological question; it's a social question, it's an economic question, a geopolitical question. And that's why I address all of those as well. So these things are all coming together at the same time. And that's another reason people kind of feel like they're being thrown off balance in every direction, because everything is changing all at once. Right?

So throughout this conversation, I feel like I've started to be able to put together a mosaic of like your view of the future through, you know, a series of different lenses, and also some spaces where you say, you know what, there's still too much uncertainty here. Is there anything you can tell me about, like your predictions for the next five to ten years that we haven't covered that, like, you're pretty confident we're going to see?

Yeah. I would say that the—the paradigm around training, skill sets, and education is going to change dramatically. And that has profound implications for the work that we do and how we go about that. One of the things I talk about is skill flux, and it's this concept that we go from this paradigm of, you know, 30 years ago, you could have a skill set that lasted you 30 years before its shelf life was obsolete. Now, you know, someone like me, I had a skill set that was, you know, ten years. It was valuable. I started off as a mobile strategist for an agency. You don't hire those anymore; it just doesn't happen. Right? You might at an enterprise level, if you're like Cisco and doing software, but not in that environment. And the skill sets that are valuable are shortening on their shelf life. And for more technical skill sets, those are arising and disappearing faster than ever. So now we're at like two and a half years for a technical skill set, and I could see that shrinking more and more and more; it's the point where many of them are rising very quickly and gone the next day, six months. To give you an example, I think coding and prompt engineering are two versions of that. So prompt engineering became something that was relevant about two years ago. I would give that maybe a five-year shelf life max before it's no longer relevant at all. And we're already seeing agents being able to take on a lot of that work. But there will also be a new skill set that you'll have to learn in order to operate in a new paradigm with new technology and new objectives. So we're going to see this exponential increase in importance and value, and a ChatGPT moment that comes in says that's not valuable anymore; that's gone, right? And that's going to have this almost like whiplash effect for us as we go along. That changes how we educate ourselves; if we're frontloading education for the first quarter of our lives, we're out of date by the time we walk out of university. And this is not a new discussion at all, but it becomes exacerbated by that. So the idea of lifelong learning, you know, very cliché, but micro-credentialing, we call it surge skilling, where it's like you're actually having to get very deep into something very, very quickly to create competitive advantage. And then you just know that this is going to be less valuable in a certain period of time. But what is valuable is being that first mover and creating value with it as fast as possible before it—before it becomes obsolete. So that's where I see education changing, where I see people shifting their focus for competitive advantage. And this—the culture of an organization changing too, because you're going to have to keep learning on the job. And the tools and the AI as you're using will have to teach you how to work with it as they change.

So I'm really glad, Ian, that that was your answer, because that was on my list of things that I wanted to talk to you about that we hadn't gotten to yet. With that in mind, this—the—the shortening time horizon of skills you mentioned, you know, it's going to have massive implications on education. What do you see as being the risks and the opportunities for the traditional education system? And also, what does it mean for like the hiring process of organizations?

Absolutely. So it completely disrupts the one-to-many broadcast model; like the idea of a teacher standing in front of a room and speaking for an hour and a half to three hours is gone. The—which is great for people like me; I was a terrible student, super neurodivergent. I can't sit in a class on a lesson for more than five minutes; I have to be engaged. So what this is going to do, it will disrupt the current model, but it will make it amenable to a much larger group of people who are not built for the more industrial, assembly-line-like education model. The challenge, though, and I don't say that with—with any malice towards teachers and educators—they are some of the most under-resourced, overtaxed, and over-expected people in the world. Then you take a look at the dynamic in the US and how hostile it is. I have so much empathy for people who choose to go into a life of service for the next generation. We need to be spending a lot more time and money in that space. And I make a comment in my keynote where I say the land budget should be as big as your technology budget, and that kind of—like, people look bug-eyed at me. Like, what? Like what we're spending—so we're spending trillions of dollars on technology.

I love that—I love that, yeah. But if we don't—like the technology's moving faster than any other sector, faster than the economy, fashion, the society is moving faster, and education's moving. And if we truly want to understand where humans play in that picture, the fact that we're investing everything we have in technology has already indicated our preference for technology over humans, so that math has to balance out a bit. We have to figure out how do we invest so much more into education? Not so much less. And until we do that, we are going to be behind the ball. We are going to have a target on our back in many ways, because if the paradigms don't change, the technology just gets better; we're going to suffer the consequences. But if we put ourselves front and center of that equation, we have the chance and the opportunity to figure that out. Right?

It's wow. Yeah, it's—as you said it, like this is not an incremental shift. This is like a complete disruption of the model from end to end, without a doubt. And even for people who live and breathe it, like it's overwhelming for me; I do this 24/7. I love it, I'm passionate about it, I'm excited about where we're going, and net-net, I'm optimistic about the long-term future, but we are all pioneers right now, whether we want to be or not. And when people—we've kind of bastardized the term pioneer; we've made it seem like, oh, it's Richard Branson on the cover of Entrepreneur magazine with his billions of dollars of success, like he was a pioneer at one point in time. But yeah, pioneers go through really hard shit, and they go to places where there's no infrastructure. They suffer the consequences of, you know, decisions that they didn't know they'd have to make. They are attacked by the environment that they're in; nature tries to kill them in a number of different ways. Yeah. And as a super resilient species, we still make a way forward. We construct the environment after we figure it out. You know, we might show up in Hawaii with no shoes on and realize, oh, crap, I'm not properly equipped for this. And then we figure a way out of that pattern. The time to go from not knowing to knowing can be really hard, painful, and challenging. But the way we thrive once we do is absolutely amazing. So I would say that we are going to have amazing things happen, but we're also going to have to encounter some really tough growing pains, individually and collectively, to get there. So if anyone is saying otherwise, it's absolutely smoke and mirrors, right?

Right. Wow. Exciting times ahead. There's one more question that I wanted to ask you that I haven't had a chance to yet, which is I wanted to ask you the—the inverse of what I just asked you, which is, you know, aside from, like, what is going to happen and what's going to disrupt us, is there anything you're hearing right now, hype-wise, technology-wise, trend-wise, that you're like, that's B.S., like that's not actually going to come to pass; we're being sold a bill of goods?

Yeah, I actually—I think the agents conversation is way overhyped. I think they are transformative. I don't know a single organization that is going to say, I'm going to let an autonomous series of agents run my enterprise that I've spent decades building, without the oversight necessary. Like, we've—we've been working with agents for years, and we've been building setups where agents will work with other agents and giving them autonomy and creating virtual environments to see what happens. And every time we let them run amok, it's frightening. Like it is absolute, like jaw-dropping. Oh my gosh, I can't believe that would have happened; so glad I didn't give them freedom to access real live data. Yeah. And that—that infrastructure needs to be built. The—there are actions that agents can do that are absolutely mind-blowing. But they're narrow, they're specific, they're structured, and they have strong guardrails. The idea that we can kind of have this almost reinforcement learning, you know, give it a million different examples, let it kind of like bang around, figure its way through approach to unleashing it in the organization does not work. Because the infrastructure is just not there yet; it hasn't caught up with the promise of the technology. So I think we're very much at the top of the hype cycle of agents. We're going to have this crash into the trough of disillusionment, which, in my opinion, is the best place for a nascent technology to be. A lot of people say less bad, but what it means is the people who are making promises who don't know what they're talking about, and let's face it, there's an enormous amount of people who are rushing to find the gold that have no business being here and making promises, they disappear because it's now—it's getting hard. You actually have to deliver. And in the trough of disillusionment, it pulls all the pundits out. And now the people who are committed to doing the work, who are there for the right reasons, they get to work and they build that infrastructure that's necessary to deliver on all those promises we were making back here. So it takes time, and I'm just kind of waiting for that to kind of implode on itself and for people to see like, oh yeah, these are very, very powerful; this is absolutely the paradigm of the future. But the future is still the future, not the present; we need to get there first. Right?

I love that answer. And I think it's so appropriate right now, given where we are in that hype cycle. So thank you for taking some of the air out of that one. That's awesome. One of the things I would—maybe I can put a finer point on is the metrics piece, and that is the expansive versus constructive. So what I was talking about earlier, as I was mentioning, a lot of teams are going to say let's—let's do more with less; let's pull back the number of resources we have and get along and have higher efficiencies, greater margins, and better stockholder returns. And a challenge that we have is we're moving into a paradigm that is going to shape what matters, and what matters, and what's valued in the work that we do is going to be different. But when what matters changes, but the metrics do not, and the incentives do not—

Yeah, yeah.

That means you run right into a paradigm that is going to push back on you and potentially hurt you as an organization. So we encourage organizations in times of change to also understand how is this going to change the incentives and the metrics that are used to measure that change as it's happening. So we're thinking more about growth metrics, metrics of innovation, metrics that are about charting the unknown versus optimizing the known. We've come from a paradigm of optimizing the known for the last 150 years; we're really good at it. The problem is how much is known about the next five years. So if we're doing 95% of our metrics on optimizing the known, 5% on exploring the unknown, that means you're already out of date. If we're starting to push more of that towards exploring and charting this unknown territory, this makes us more prepared for what's going to be coming. This gives us the opportunity to think about innovation quotient and knowledge diffusion across the organization, building the structures that will make you resilient in this future paradigm. Because right now, optimization—organizations at scale, by definition, require some calcification of the organization; it needs to be rigid in some ways in order to be efficient. And rigidness against an oncoming wave is a recipe for disaster. So that's one of the things we encourage organizations to think about. And we get very deep into what is that metric, what matters for you? How is it specific to your context? What are the things that you measure? How do you actually do that work? And when that clarity is there, all of a sudden it goes from, well, we don't know what the future brings, to at least we know how to move in that direction. Right.

And with respect to the fact that I'm sure there's lots of different metrics for different organizations, is the answer to, like, just move to a new set of hard metrics or get more comfortable with the notion that we need to be flexible and measure things with a little bit more flexibility than we have in the past?

Absolutely. That—that's typically the first step. You never want to shift entirely because you kind of want to leave what's working, working. So we don't say you've measured this way, don't do that anymore, but at least a portion of the work that's being done needs to be done in this forward-facing way. And that type of work needs to be measured differently. Because if you were measuring, for example, a lot of teams, a lot of tiger teams, a lot of innovation teams are measured by ROI on their first run, which is mind-boggling to me. Okay, right. You're going to have impact on margin the first time you touch ChatGPT? No, that doesn't happen. And yeah, I think that's almost a bad example because that's just ludicrous on all levels. But if you're thinking about scale, efficiency, margin impacts on things that are by definition going to require investment and time, you're already impeding the work that is going to help you explore unknown territory. Yeah.

Now it's—it's super interesting. And yeah, I'm sure we could talk for another hour just on—just on that. With that in mind, though, you know, I did want to say a big thank you for joining me today. This has been super, super interesting. And it's honestly been a real treat. I talk to a lot of people in this space, and I'm just continuously blown away by the breadth and the depth of insights that you have in this space. So I wanted to say a big thank you.

Thanks, Jeff. It's been an honor to join you; I really enjoyed it.