AI Can’t Do Your Job. So Why Are Companies Laying Off Employees?

0:00Over the past week alone, there have been multiple announcements about mass layoffs and organizational restructuring from Amazon's announcement about laying off 14 ,000 corporate workers to the CEO of YouTube stating that employees can voluntarily exit the company and they will be given a severance package to a more soft restructuring where companies are slowing down the rate of hiring. Now, a lot of the press and media narratives is that this is AI, the beginning of AI starting to take jobs and automating workflows. And that's not really true. AI is involved in these layoff decisions, but not for the reasons people think.

0:38Because whether AI can do your job or not, and believe me, it probably can't, is likely irrelevant to the fate of the job itself. I'm going to explain what's happening, what we can expect going forward, and most importantly, what we should be doing in this moment to stay ahead of it. So these announcements seem to keep coming. It feels like the workforce just can't catch a break. And for some of these layoffs, it is standard cost-cutting, standard productivity moves, or some course-correcting for pandemic overhiring. That's some of it. But the first way that AI is playing a role is that companies are starting to reallocate capital away from labor and towards AI infrastructure.

1:22So investments in artificial intelligence are expensive. Billions of dollars worth of investments, to put it bluntly, from the energy costs to the chips needed to train artificial intelligence systems. This costs a lot of money. So it seems like companies are starting to place a bet on AI infrastructure, knowing that AI is becoming a foundational technology, that all companies will be building on top of it, and that money has to come from somewhere. So that is some of the organizational restructuring decisions. But the main reason AI is playing a role in these decisions is much more nuanced. First and foremost, companies are not automating roles because AI can complete those workflows.

2:06AI in its current form is not capable of automating thousands of workflows. It is not capable of automation at scale at all. I don't care how many leaked memos from CEOs we hear about how AI is transforming their organization. It can do all these jobs. It can't. It still makes a lot of mistakes. It's very challenging for this technology to complete a workflow from start to finish. So that is not what is happening here. Companies are slimming down and becoming more nimble and much more adaptive and much more flexible to prepare for the disruption that they know AI is going to bring. In other words, it's like an anticipatory restructuring.

2:48Organizational leaders aren't asking, can AI do the jobs of these people? Because the technology cannot. They're saying, we can't afford to move slowly when AI changes the rules. By cutting jobs and consolidating teams, they're trying to build organizations that are adaptive, flexible, entrepreneurial, companies that can experiment quickly and reallocate resources fast and pivot the moment a new AI capability changes the game and changes the landscape. And here's what I mean by this. Think about all of the companies that were disrupted by the internet. And we talked about this in Ajay's episode and in Manu's episode, that the companies that were leading pre-internet, almost all of them were off the top 20 list post-internet.

3:38With every new general purpose technology, like the internet and electricity and now AI, there will be winners and there will be losers. And usually that disruption comes from the horizon, the edge of the market, where big companies, it's hard for them to not only see that disruption coming, but it's hard for them to then pivot and adapt and prevent that disruption. So in a sense, companies are trying to internalize the disruption now rather than externalize that disruption or outsource it by someone else. So instead of waiting for an AI native startup to eat their lunch or to pivot faster than them, they're trying to make themselves behave like a startup.

4:19Lean, fast, experimental, unencumbered by bureaucracy. And we can see this in the statements these CEOs are making. So Amazon's senior vice president of People's Experience in Technology stated in a recent blog post, this generation of AI is the most transformative technology we've seen since the internet. And it's enabling companies to innovate much faster than ever before. We're convinced that we need to be organized more leanly with fewer layers and more ownership to move as quickly as possible for our customers and our business. Amazon knows that Amazon was invented because of the internet.

4:55So they're looking around to think, who could be the next Amazon in the AI first era? If we don't start to behave like a startup ourselves, we will be disrupted by what could be the next Amazon. And the CEO of YouTube states something similar. AI is the next frontier for YouTube. Looking into the future, the next frontier for YouTube is AI, which has the potential to transform every part of the platform. We need to set ourselves up to make the most of this opportunity. The CEO of YouTube isn't planning for a future where creators are now using AI-generated content. No, no, no. That's not the disruption they're thinking about.

5:33They're thinking about what does content even mean in the era of artificial intelligence? In other words, they're thinking, okay, in 1995, what should cable news have been asking about what the internet would mean for how people share and receive content? A fundamentally different type of organization was born because of this technology. So that's what these companies are doing. They are starting to internalize their own disruption, slim down, become much more flexible so they can be prepared for anything and they can innovate and chase the opportunities like a startup. And it doesn't mean that they'll be able to do it, but that's what they're positioning for.

6:18And so when it comes to jobs, they're basically making two big bets. The first thing they're betting on is AI's capability, that AI will eventually automate more workflows than it can do today. So investing in it now would give them an early start. But the second bet that they're making, which is the bigger bet, but the more important one, is that the workflow itself could just be irrelevant anyways. So even though AI probably can't do that full job today and probably not even in the next couple of years, the workflow itself might not even matter anymore in an AI driven economy. So let's use AI as a forcing function now to prepare for whatever it is that we will be optimizing for going forward.

6:59And that's why the current debate around can AI do your job is so misleading and potentially dangerous because that's not the question companies are asking. AI probably can't do your job as it stands today, or it can with a lot of mistakes and is probably going to be more work for the person who has to correct those mistakes anyways. Most people are looking at AI through the lens of technical performance. It makes mistakes. It has its limitations. It struggles with basic things. and then they make the incorrect assumption that the job is safe, but that misses the strategic reality entirely. Companies are not worried about whether AI can do a specific job today.

7:41They're worried about whether your workflow would even matter in the systems that are coming next. And they're making a bet that either AI will soon be able to be good enough to do that task, or maybe that entire function and all of those tasks within that workflow are irrelevant in the next version of their business. So when you use AI at work and it makes mistakes, one, it could mean that maybe we're not using it properly, but it also might not even matter because the company you work for, it probably isn't optimizing around AI's current accuracy. And it's really important that we get that nuance.

8:17The same way we feel quite a bit of uncertainty about jobs going forward and the state of the workforce, companies also feel uncertain about the longevity of their organization. They are slimming down as fast as possible and getting into startup mode. So this means we're going to hear about more mass layoffs coming down the pipeline. It also means we're going to hear more incorrect critiques that companies are prematurely firing people because AI can't do those roles. But now we know that that's not what's actually going on. They don't actually care if AI can do the role. And if you do hear a CEO say that AI can do the role, then all the best to that CEO because most leaders should know that AI isn't capable of full automation at this point in time.

9:01So what should we be doing in this moment? And that is what's most important to understand. First and foremost, you need to learn how to work with artificial intelligence and not to fear it. And I know it kind of sounds scary with all of these moves, but understanding how to work with AI, you have to consider it like understanding how to work with a computer, it's going to become the standard platform that all workflows are built on top of, that all companies are rebuilt on top of as well. So the more you can understand how to leverage this technology, the better equipped you are for your entire career, regardless of where that is and what it specifically is you're doing.

9:37But with respect to the company that you're working at today, you want to try to make yourself as indispensable and as valuable as possible. So knowing companies are positioning themselves for this AI-first future, the more you understand how AI could support you in your role and the company more broadly, the bigger the asset you're going to be to the organization. But even more broadly, we need to start building all of the skills that are required to thrive in the AI-first era, regardless of where we are executing and performing these skills. Skills we've talked about in a lot of the previous episodes, like judgment.

10:13What questions do you know to ask these supercomputers? Are those the best questions to ask them in the first place? Do you know how to evaluate their responses? Because that is where the future is going. We are moving towards a world where we direct AI systems. And consider it like, if you had five interns today, what would you ask them to do? Start thinking about AI in that way. And how would you evaluate their work? So that's a really important judgment skill. And it's a lot harder than people think. People think you can just outsource your thinking to AI. And you can, but you will be out-competed by doing that because AI should technically raise the bar on what we have to bring to the table because we're asking questions we couldn't ask before because now we have access to these systems.

10:52The same way a computer allowed people to look at a lot more data than they could before when they just had paper spreadsheets so they could analyze new material. AI will allow you to look at things differently. Do you know what you should be looking at as a result. So that's kind of the judgment skill and the critical thinking skill, building communication skills, understanding how these AI systems even work, how do they make the predictions that they do, what data have they been trained on. But then some of the skills, and people call them soft skills, but I think they're actually in some ways harder to learn.

11:26Adaptability, flexibility, and learning how to learn. Everything we've been talking about, and especially on this podcast, involves learning to do things differently, learning how to work with these systems, learning how to pivot as roles start to adapt and as we start to move more fluidly in roles, which I'll talk about in a second. And learning is actually kind of tricky, right? Because most of us don't know, how do we learn? Where do we learn best? In what type of manner can we learn most easily? These are the types of questions we need to be asking ourselves now. And here's why learning is so vital.

11:58The restructuring companies are going through now isn't going to be a one and done change. When you think about artificial intelligence, this technology is going to continue to learn more tricks over time. What it can do today will be different in 24 months, in 48 months, and so forth. So companies are going to stop hiring for as many full-time roles as they had traditionally because you don't really even know what those roles are going to entail in 24 months or in 48 months. In other words, we are moving out of the era of this kind of full-time employment where your job is static and where you move vertically up this career ladder based on predictable steps ahead and towards a much more fluid workforce and fluid workplace that is much more about skills than it is jobs.

12:51For instance, you might be a financial analyst today, but in a couple years you might be directing AI systems to do the analytical work and you're making the judgment calls based on that work. So are you now a financial strategist? And if you don't have the judgment skills, maybe somebody who has stronger judgment skills, but weaker analytic skills, but stronger judgment skills than you, is better equipped for those tasks, even if they have less experience than you. So, and we talked about this in Ajay's episode. It's going to be skills over jobs. and skills over experience. And this is an entirely different fabric of the workforce, one where we work more independently, one where we move much more fluidly between different types of roles and projects, and one where there's continual change.

13:38And I call this the rise of the independence era, where the dominant fabric of the workforce over the next couple years will be much more independent workers and much more entrepreneurial. So we'll have to start thinking of ourselves as entrepreneurs that offer a bundle of skills and the propensity and the ability to learn and pivot and adapt. That is the dominant fabric of the workforce going forward. That idea that we learn, we work and move up a ladder, the career ladder in and of itself, that's going away. We're moving away from job titles and more towards a dynamic portfolio of skills. And we're going to work across multiple projects for multiple companies at once.

14:20This is a really big shift to the workforce. And it comes with a lot of uncertainty, right? On the one hand, even just social security. A lot of people have their insurance attached to the job that they do. There's going to be big questions around that and questions that we need to start thinking about now. And I've raised these questions quite frequently on my own social platforms and rooms of policymakers on my sub stack. These are issues that we can see quite clearly right now. But then there's also the idea that a lot of people don't even want to be an entrepreneur or an independent worker.

14:54There is a lot of safety and security when you can see work as something much more stable. And that chapter of the workforce is closing. And that's really tough. But the more we can start to anticipate these changes, the more we can prepare for them. The more we can start to isolate and understand what skills do we have today, what ones should we be building. We can more easily start to think of ourselves as these mini organizations that come with a bundle of skills. So the more we confront this, the easier it's going to be. And there are a lot of people that are already opting into this independent era of work anyways.

15:28So if you are a creator, if you are a freelancer, or there's a lot of people that work nine to five and then they have their side hustle or they work on a project that they love from five to nine on the week or on the weekends or after work. Those types of projects and kind of different freelance opportunities, that's going to become much more commonplace. And it's possible that eventually this workforce disruption feels a lot more smooth. And the idea that we're more independent workers starts to feel like the new norm for work. but it's going to take some adjusting to get there. The same way what we do today and what we consider work today would have been completely unrecognizable to people over 100 years ago.

16:09We have to expect parts of the workforce to be entirely unrecognizable because of this technology. So in other words, jobs are going away, but work isn't. Job titles are going away, but the skills and the value we bring isn't. We're going to continue to cover the future of work over and over on iCropQuestions. We're going to continue to track the moves companies are making, the skills that are needed, the skills that we need to be building. We're going to continue to bring on experts. I would recommend watching the AI Economist episode with Ajay Agarwal, where he talks in depth about the skills we need to build and how to build them.

16:45I would recommend watching Alexander Manu's episode where he talks about these more philosophical changes that are coming to the workforce and how we can start to understand our identity going forward as it relates to work. I write about the future of work in the weekly newsletter on Subsac, and we're going to link that newsletter below. And we want to hear your specific questions about your job, about your company, about the industry that you're in, and we'll answer them directly on the show. And you can submit these questions via our website, IGQ with Sinead Bovell.com. And we will be navigating these dynamics with you every step of the way.

17:19So thanks so much for tuning in this week. I look forward to seeing you at the next episode.