Job Market 2026: Why Everyone Is Getting Laid Off—And How to Be the Exception — Silicon Valley Girl Podcast

Marina Mogilko: So this morning I opened my phone and it's the same story again and again. First, we heard a few days ago that Jack Dorsey's blog just laid off almost half of his stuff and says AI is the reason. Just recently, Atlassian cuts 1,600 people to self-fund investments in AI. Your feed, my feed is full of "AI took my job" posts. Is this the beginning of a jobless future or are companies just slapping AI on old school layoffs to make them sound innovative?

Today we're going to look at the data, the companies, and the people behind the headlines. And I will be joined by someone who actually knows the numbers. I actually use a report she creates every two years. It's called the Jobs Report by the World Economic Forum. She knows all the numbers and she knows exactly what's happening on the job market. Are these layoffs AI? Should we be afraid? Should we be scared? Or are companies just using the word AI to justify layoffs? By the end, you'll know which jobs are actually at risk, what AI native really means, and what you can practically do in the next 90 days to future proof your career, even if you're not a tech person. I'm not a tech person, but I'm using so much AI. I'm going to talk about this.

So, what's going on on the market? Let's start with the headline that broke the internet. Jack Dorsey and Block laid off 4,000 people. Block is the company behind Square, Cash App, and Afterpay. They announced that they're cutting 40% of its workforce, more than 4,000 people, bringing headcount down to under 6,000. Jack Dorsey didn't say, "We had a terrible quarter." He didn't blame a recession. He explicitly pointed at intelligent tools and AI agents. In his letter and interviews, he says smaller, flatter teams can now do the same work, that AI fundamentally changes what it means to build and run a company, and that most firms will inevitably follow this path. Media from CNN to Forbes jumped on this story as one of the clearest examples of a CEO saying, "Will we lay people off because of AI?"

Okay, I had to talk to someone who's a professional in this. I spoke with Saadia Zahidi, managing director of World Economic Forum. I'm going to share what they told me, which is that a top person on jobs at the organization feels that a lot of firms are sort of maybe using this moment of concern around artificial intelligence to essentially take care of the issue of overhiring that happened three years ago. And now is the time to change that. It's sort of a somewhat convenient moment to use this as an excuse.

Saadia Zahidi: From inside the room, the message is pretty blunt. Yes, AI is real, but it's also a perfect excuse to fix the overhiring that happened during the last boom.

Marina Mogilko: Now, let's talk about Atlassian and the AI restructuring narrative. Right after Block, Atlassian announced it would cut about 10% of its global workforce, roughly around 1,600 people. Atlassian says it's self-funding bigger bets on AI and enterprise products. The CEO says that AI is changing the mix of skills we need, the number of roles required in certain areas, and he says we can't really pretend AI has nothing to do with it. And it's not just these two companies. Across 2025 and early 2026, companies from Amazon and Microsoft to media and fintech startups have cited AI in their layoff announcements. In the US alone, employers explicitly blamed AI for tens of thousands of job cuts in 2025. A huge jump from the years before. Sometimes that's real automation. Sometimes it's just smart branding for cost cutting.

Okay, let's move from buzzwords and press releases to actual data. In early March, Anthropic, the company behind Claude, released one of the most detailed studies so far on AI and jobs. And I absolutely loved it. Thank you so much, Anthropic, for doing that. It's called "Labor Market Impacts of AI: A New Measure and Early Evidence." So instead of asking in theory, could AI do this task? They looked at observed exposure where workers are already using tools like Claude to do their jobs right now. And here is what they found.

For computer and math jobs—hello, my education. Software engineers, data scientists—large language models could touch something like 90 plus percent of tasks. But in practice, workers report using AI on only a fraction of them. So far, the jobs with the highest AI exposure today are white collar roles: business and finance, computer science, law, computer service. Comment down below if your job is listed here or your education is listed here. I'm mathematics and economics by education. Oh, and of course, office administration.

There are some safe jobs. About 30% of workers have basically zero AI exposure right now. Cooks, bartenders, mechanics, cleaners, many trades, and in-person service jobs. And here's the subtle but important point. They don't see a big spike in unemployment yet for most AI exposed jobs. What they do see is a slowdown in hiring into those roles, especially for younger workers. So the people in those jobs aren't losing them overnight, but it's getting harder for new people to get into those fields in the first place.

Let me tell you my personal story. I totally see this trend. I run three YouTube channels, two Instagrams, an email newsletter, a lot of things. My team uses AI every single day for research, for scripts, for translations, for thumbnails. And we've been doing that for about 18 months now. And in the past three weeks, we deployed Claude projects for almost every single social media that we run. And that sped up things immensely. I don't know if you noticed, but we started posting two videos on this channel instead of one. Production just sped up. What one person does has changed completely. My producers are no longer spending hours on research. They are not writing first drafts from scratch. That time went into something else like producing more episodes, quality control, judgment calls into the stuff that actually requires a human. And honestly, that shift from doing the work to directing it changes what you're worth.

Around four weeks ago, I asked my chief of staff and COO to hire someone who would help me with scripts for brand deals. Because you need to look at what the brand sends you, then it has to be a personal story, and there are so many different points you need to connect for that. We started testing some people and we found someone who's genius, but £1,000 for a script for a 60-second video. I'm like, expensive. And I trained a Claude project and we completely eliminated the need to hire someone for just scripts. Did that scare me? Yes. The thing is it is happening. It kind of makes me sad, but it kind of wants me to help you guys with the next steps because it is happening.

And this is basically the story that this Anthropic report is actually telling. The jobs aren't disappearing yet. The tasks inside those jobs are being reshuffled. One person can handle more. You don't need more people. And the people who understand which of their tasks are being reshuffled and move proactively are the ones who come out ahead.

Saadia explains this with a very simple picture. Imagine the whole global workforce is just 100 people.

Saadia Zahidi: Let's pretend the entire global workforce is 100 people. About 50-something of them would need rapid reskilling by 2030. About two-thirds of those could be reskilled within their current role. That's what employers are telling us. And about one-third of them would have to be reskilled or upskilled and redeployed into a different role inside the organization. However, there's about 11 people in this overall 100-person workforce that wouldn't necessarily have an easy place to be reskilled to. And I think that's where the tension comes in.

Marina Mogilko: A lot of the change could probably be handled where people are currently employed. But there is going to be a significant share of the global workforce that will not find a role inside their own industry and that will need to move to a completely different industry. And that is where basic human connection and your social networks and your professional networks—that's where all of that comes in.

Saadia Zahidi: Of those 11 people, which industries are they in?

Marina Mogilko: Oh, multiple industries. We do take a look at what are the declining roles, what are the growing roles. When it comes to the declining side, I mean you wouldn't be surprised to know that it includes people who are administrative assistants. It's people who are in some parts of customer service that is now getting digitally automated. But there's also a lot of growth, right? The highest growth sectors include agriculture, include education. These are all things that the world population still very much needs. There's a huge shortage of teachers around the world that is still very much needed. And it doesn't mean that a profession such as that one is getting replaced by technology. So I do want to present the full picture beyond the ones that are getting displaced.

Saadia Zahidi: So the real danger isn't AI takes everyone's jobs. It's those 11 people out of every 100 who don't have a clear place to go. So even if today's layoffs are sometimes AI-washed for shareholders—it looks great to shareholders, we're optimizing, we're investing in the future—sometimes that's true, sometimes it's just a glossy label on old school cost cuts. But under the PR, the deeper trend is still real and unavoidable. Routine rules-based office work is being eaten by software and AI. New jobs are appearing in tech, green industries, healthcare, education, and they demand a different mix of skills. The premium is rising on people who can combine human skills, AI tools, and domain expertise.

Marina Mogilko: Let's talk about safer zones. Safer doesn't mean perfectly safe forever. But today, jobs with low AI exposure tend to be reality-native jobs where you physically fix, cook, build, or care for things. Mechanics, electricians, plumbers. You've probably seen those viral reels where the plumber is the new millionaire. Chefs, cleaners, delivery, construction, high-touch human roles where your main value is emotional intelligence, complex human judgment. A strategist inside a company or early childhood education, many healthcare roles, social and community work. These are much harder to fully automate because they combine physical presence, messy real-world environments, and human involvement.

Let me give you a useful framework, a way to figure out exactly where you personally stand. Every job has two layers. Layer one is the task layer, the repeatable rule-based work. Write this email, fill this report, answer this ticket, schedule this meeting, process the claim. That's layer one. You're below AI here. Layer two is above AI and this is the judgment layer. The stuff that requires context, relationships, feel, intuition, experience. AI is eating layer one fast everywhere across every industry, not just in tech. In law, in marketing, accounting, and medicine and customer support.

The Anthropic report found that for white collar roles, AI can technically touch over 90% of tasks. In practice, workers are using it maybe 30 to 40% right now, but that number is growing every six months. So the question is not, is my job at risk. The question is, what percentage of my day is layer one versus layer two? If your answer is 80% layer one, meaning most of what you do is routine, templated, rule-based, you have a real problem. Not because you're going to be fired tomorrow, but because every year that passes, your layer one work gets cheaper and the company doing your annual review knows that. If your answer is 80% layer two—judgment, relationship, strategy, creative decisions—you're actually in a stronger position than you were two years ago. Because again, AI is handling the layer one work for you, which means you can do more layer two work in the same hours.

Which brings us to the real question. If you want to stay ahead of this curve, what should you actually learn? I asked Saadia what other skills every worker needs to develop today.

Saadia Zahidi: I would say the human skills have just become even more important. So, oddly enough, in a highly technologically driven world, it is the human skills that have become more important than ever before. And that's very clear from every two years we put out this list of the top 10 skills, and that's basically what comes through very loud and clear through that feature of jobs survey. That's one piece.

There is some element of that which is also about being able to work closely with technology, but it's a couple of the top 10. Most of the skills in the top 10 relate to creativity, empathy, interpersonal interactions, leadership skills, social influence, self-management, being able to regulate your own self, especially in the midst of so much change. So those are the kinds of things that are rising to the top.

Third point, unfortunately, those are not the skills that most employers test for, right? So most employers say that's what they want, but when you get into an interview process, that is rarely ever the set of things. Unless people are doing it anecdotally, that is rarely ever the set of things that they're looking for.

Marina Mogilko: So the paradox of the AI era is: the more powerful the technology becomes, the more valuable the human skills become. And if you look across Saadia's work, the Future of Jobs Report, and the new AI research, the people who thrive tend to have a three-part skill set.

First, human skills: creativity, communication, empathy, leadership, social influence, self-management, the ability to handle work with other people.

Second, AI skills. Not building models from scratch, but using tools like Claude, ChatGPT, Copilot, and others to write first drafts, summarize, brainstorm, generate ideas, analyze data. Knowing how to design good prompts and workflows so that AI does the boring part.

Third, domain skills. Real expertise in something. Marketing, law, finance, medicine, design, trades. The context you need to judge whether AI's answer makes sense and to turn it into real decisions. By the way, being AI-native doesn't mean you work at a startup in Silicon Valley. It means that whatever you do—HR, sales, design, small business—your default is: what can I offload to AI so I can spend more time on the uniquely human parts of this job?

The challenge is that most schools and companies still don't explicitly teach those skills. So where do you actually practice them?

Saadia Zahidi: I think education systems are starting to change and recognize that the previous education system which was deeply individualistic, competitive, right? If you got the number one position in your class, well, someone else had to be number two. That's the old model. But I think a lot of education systems are starting to change, certainly from what we heard from some of the universities here, because they recognize the value of collaboration. They recognize the value of group work. And I would, you know, if there's one piece of advice I would give to anybody who is currently a student, take on some of those group projects, work with other people, see how you can negotiate, work together, collaborate. And you'll find that working with others, you're going to learn so much more, and it'll prepare you for the workforce because that's what most of our workforces are.

Marina Mogilko: So if you're a student or early in your career, the fastest way to build future proof skills isn't just another online course. It's projects with other people, group assignments, hackathons, side hustles, startups, volunteering, anywhere you have to negotiate, coordinate, and ship something together. To make this practical, here's a simple way to start.

In the next 30 days, pick one AI tool and use it every day for your current work or studies—for writing, summarizing, analysis, or planning.

In the next 60 days, ship one small AI-powered improvement. An automated report, a smarter template, a content generator. Perplexity Computer is blowing my mind right now with content generation. Something that serves you real time.

In the next 90 days, pick one human skill from Saadia's list. Communication, negotiation, leadership, self-management. And deliberately practice it in a real project with other people. If you do just that, you're already ahead of most people reacting to AI from fear instead of design.

So, where does it all leave us? Yes, some big companies are hiding behind AI when they fire people. It looks great to shareholders: we're optimizing, we're investing in the future. Sometimes that's true. Sometimes it's just a glossy label on old school cost cuts. But under the PR, the deeper trend is still real and unavoidable. Routine rules-based office work is being eaten by software and AI. New jobs are appearing in tech, green industries, healthcare, education, and they demand a different mix of skills. The premium is rising on people who can combine human skills, AI tools, and domain expertise.

We asked Saadia what she would say to a young person who feels anxious seeing AI, geopolitical tensions, and automation all at once.

Saadia Zahidi: All of these are potential trends. Some of these are trends that are highly likely to come true. Yes. But throughout the last 20 years that I've been in the workplace, we have had these moments before. There has been geopolitical tension before. There have been wars before. There have been technological changes before. A lot has happened. There's been the great financial crisis. So there have been many disruptions. There's been COVID. So I think we should bear in mind that that hasn't dented employment yet. That has required us to be resilient and to adapt, and those are certainly skills that young people should develop. But I'm very—I remain always very optimistic and hopeful about the future. And that would maybe be my one piece of advice. Invest in yourself and stay optimistic.

Marina Mogilko: AI will absolutely change what jobs exist and how we do them. But it doesn't get to write the whole story. Governments, companies, universities, and you still have choices. You can control whether a CEO decides to announce AI layoffs. You can control whether you become the kind of person whose job is easy to narrate away or the kind of person who knows how to use AI, lead people, and create value in the new system.

If you treat AI as a thing that's going to take away your job, you'll spend the next few years really scared. If you treat it as new electricity for your career, your real job becomes to design how you're going to use it. And that job is only just beginning.