Daybreak - AI probably can't do your job yet. But it might get you fired anyway

Episode Date: February 4, 2026

Amazon fired 16,000 workers last month. Oracle is set to cut up to 30,000 more.Tech layoffs have increasingly been attributed to AI. But Oxford Economics found something strange: there's no macroeconomic data showing AI is actually replacing jobs or boosting productivity. In fact, output per worker is slowing, not accelerating.So what's really happening? Host Rachel Varghese breaks it down.Daybreak is produced from the newsroom of The Ken, India’s first subscriber-only business news platform. Subscribe for more exclusive, deeply-reported, and analytical business stories.

Hi, this is Rohan Dharma Kumar. If you've heard any of the Ken's podcasts, you've probably heard me, my interruptions, my analogies, and my contrarian takes on most topics. And you might rightly be wondering why am I interrupting this episode too. It's for a special announcement. For the last few months, I and Sita Raman Ganeshan, my colleague and the Ken's deputy editor, have been working on an ambitious new podcast. It's called Intermission.

We want to tell the secret sauce stories of India's greatest companies. Stories of how they were born, how they fought to survive, how they build their organizations and culture, how they manage to innovate and thrive over decades, and most importantly, how they're poised today. To do that, Sita and I have been reading books, poring over reports, going through financial statements, digging up archives, and talking to dozens of people. And if that wasn't enough, we also decided to throw in video into the mix. Yes, you heard that right. Intermission has also had to find its footing in the world of multi-camera shoots in professional studios, laborious editing, and extensive post-production.

Sita and I are still reeling from the intensity of our first studio recording. Intermission launches on March 23rd. To get alert, as soon as we release our first video. episode, please follow intermission on Spotify and Apple Podcast or subscribe to the Ken's YouTube channel. You can find all of the links at the ken.com slash I am. With that, back to your episode. Just last month, a certain number of Amazon workers received an accidental email. The email prematurely confirmed something they'd been expecting for weeks. Another mask layoff was in order.

The email landed on a Tuesday evening and by Wednesday morning, the company had verified that the contents of the email was true. This time, it would cut 16,000 jobs. That's after, it had already cut 14,000 jobs just last October. Then, this week, Oracle announced that it would be cutting 20,000 to 30,000 jobs. Now, since last year, tech companies, everywhere have been on a layoffs pre. The most recent situations also include layoffs at companies like Pinterest, the popular

pinboard app, Expedia, the travel booking company, and even Dow, a chemical company. The cost behind these layoffs? You may have guessed it. It's AI. But not in the way you may be thinking. I'm sure we're all familiar with the panic around AI replacing jobs, how that one system could do the work of 100 people leaving. thousands unemployed. Now, yes, of course, those possibilities do remain. But there is still a gap

between that possibility and reality. See, in some particular roles, especially in like tech, customer service, marketing and graphic design, companies are seeing widespread adoption of AI. But experts say that we're still not at a stage where AI is replacing workers and mass. In fact, an Oxford economics research briefing from last month show that there's a lot of anecdotal data about AI-based job displacement. But actually, the macroeconomic data does not suggest that there is any substantial shift in employment because of AI. See, Oxford suggests a very simple litmus test to see if the AI revolution is actually giving results in terms of productivity. Think about it this way.

If people are being laid off so that AI can do what they used to do, then the output per remaining worker should be accelerating. But that just isn't the case. So their summarization was, well, pretty bleak. Apparently, this layoff trend and the conversation surrounding it points to a pretty cynical corporate strategy. The Oxford researchers actually suspect that some firms are using AI to paint layoffs as a good news. story instead of bad news. So, if these layoffs aren't happening because AI is going to take over, why are they happening?

Welcome to Daybreak, a business podcast from the Ken. I'm your host, Rachel Verkes, and every day of the week, my co-host, Nikita Sharma and I will bring you one new story that is worth understanding and worth your time. Today is Wednesday, the 4th of February. One of the most telling hints that AI isn't actually replacing workers en masse is what an employee named N. Lee Plumb had to say. Plum worked at Amazon for eight years before getting laid off last month. He told Associated Press that when he got his notice, he knew it wasn't because he wasn't adapting to AI. Now, Plum had actually been his team's head of AI enablement. In fact, he was so on board

with the company's AI mandates, he was recognized within the company as one of its top AI users. So this is Plum's hunch on what the layoff was really about. AI has to drive a return on investment. So when you, as a company, reduce headcount, you've demonstrated efficiency, you attract more capital and the share price goes up. You could have potentially just been bloated in the first place. But when you reduce the headcount and attribute it to AI, you get yourself a value story. It's a hunch confirmed by experts.

like Cornell Business Professor Karan Girotra. He pointed out in the AP article that AI's benefits mostly go to individual workers, not the company. Now, we'll get into this in detail pretty soon. As we all know, the AI bubble chatter has been going on for a while. There are a lot of questions about how much investment has gone into the technology and how it's not really seen any massive returns over the last year or so. Major investors like SoftBank have pulled out of big companies like Nvidia, while individual-starred investors like Michael Bury have also bet on the AI bubble bursting.

So when a company has to prove gains from AI, the easiest thing to do is to reduce staff headcount and build an image where AI is being implemented across the board and say that it's driving productivity, output, and ultimately generating revenue. But of course, like I mentioned, mentioned earlier, that doesn't really seem to be the case. We are not seeing any human labor replaced by hyperproductive AI labor. Actually, if anything, it's the opposite. The Oxford research observes that productivity growth has actually decelerated, which now I should clarify,

the deceleration is not because of AI. According to the research, it's more in keeping with trends that align with cyclical economic behaviors. And the current data suggests that while productivity gains due to AI will likely show up over time, right now AI use remains experimental in nature and isn't replacing workers on a major scale yet. Plus, AI isn't so easy or cheap to implement, especially if you already have a sizable organization that's also entirely untrained in the use of this new technology. More on this in the next segment. Turns out, a lot of the layoffs and hiring phrases are less about what AI can do right now.

It's an anticipation of what AI might do later. That's the conclusion from a Harvard Business Review survey from December last year. It also found two reasons why a combination of headcount reduction and AI implementation isn't making much of a difference in terms of productivity. First, because AI typically performs specific tasks, not entirely. jobs. Here's an example they shared. Nobel laureate Geoffrey Hinton said in 2019 that AI could outperform radiologists at reading scans and spotting irregularities. Flash forward to now and if anything, there's actually a shortage of radiologists because well reading images of scans isn't the

only thing they do. Second, it's actually pretty time-consuming to do the work that identifies which jobs can be replaced with AI. To understand the real on-ground impact of having Gen AI perform tasks, you need to be conducting well-controlled and well-monitored experiments. And not all organizations have the resources to do that. So far, there's some evidence to suggest at the current stage that there are proficiency gains on an individual level. Basically, people are able to finish certain tasks faster and more efficiently. But translating this individual gain substantially at an organizational level is challenging. Also, while executives harp on about the boons of AI-enabled workforces, the workforces

themselves have a different story to tell. A recent survey by Wall Street Journal showed that employees felt AI wasn't saving them all that much time, about two hours or less at the most. and some are even feeling overwhelmed by the pressure to integrate new tools into their current processes. About 40% of the respondents even said that they'd be absolutely fine never using AI again. Overall, the mood around AI at an employee level is more anxious and less excited. And all these layoffs, whether the real reason be AI or otherwise, isn't really helping productivity either. Stay tuned.

The last Amazon mass layoff in October 2025 was attributed directly to AI at first. Current CEO Andy Jesse originally stated that he was pushing to reduce Amazon's total corporate workforce as it gets efficiency gains from using AI extensively across the company. The story around the latest Katho is a little different. The memoirs shared by the company talk more about restructuring and reducing bureaucracy. Some experts think that Amazon wants to reallocate resources into building AI infrastructure like data centers and hiring people who will actually build that infrastructure as well. Others suspect Amazon could also just be trimming a workforce that got bloated during its pandemic hiring spree.

Oracle is also doing something similar. The company has come under intense scrutiny for having its AI ambitions tied very closely to Open AI, which, as we covered in an episode a couple weeks ago, is still burning cash and not making any profit. Oracle has taken on projects to build data centers for Open AI, and this project is estimated to cost over $150 billion. Now, after certain banks pulled out of financing this build-out, the layoffs became a way for Oracle to free up about $8 to $10 billion in cash flow. So, you see how these job cuts are being driven by AI. but not because these workers are going to be replaced by machines. It's because the machines that might actually be able to replace them still need to be built.

But this particular continued narrative that AI is replacing jobs is starting to crack. And it's having a very negative effect on the remaining workforce. The HBR study found that many employees view AI with cynicism. They fear their jobs will be the next to go, so they don't put. in the effort to learn AI tools. And then, this cynicism also bleeds into their lives as consumers. They'd rather not engage with AI-powered products and services. Other companies are also seeing the direct results of making headcount transitions

based on lofty predictions instead of ground reality. Ferrester Research, an advisory firm, reported in a study that 55% of employers regret letting their employees go, because when they realize AI-Cubees, can't fully take over those jobs, they face a choice. Rehire at the same salaries or hire offshore at cheaper rates. Of course, profit-minded companies are likely to take the second route. All of this leads to mid-level employees being laid off while entry-level jobs are also being made redundant because the mid-level employees that do remain have to use AI to finish tasks that an entry-level hire would normally have done. It's creating an unfortunate

paradox. The workforce that remains behind is under-trained in AI use, mainly because companies don't invest enough in training, while Genzi, which is naturally the most AI-proficient generation, are not being hired at the entry level anymore because those jobs don't exist. This cycle is also creating a new class of disengaged workers, which makes up about 25% of the workforce according to the Forester study. These employees don't believe their companies deserve their effort because they're witnessing a clear trend. Co-workers are getting fired for AI tools that never arrive. Junior positions are disappearing. There's no fresh talent and outsourcing is getting rebranded as progress. The research shows that

when a quarter of employees stop putting in that extra bit of effort, no AI system can make up for that kind of drop in output. What we have here now is a vicious cycle, one where there really seems to be no winning, at least not right now. Companies may be able to look like they're at the forefront of new technology for the sake of investors. But still, large swaths of people are left in the large, and even the remaining workforce is slowly heading towards less productivity. And if AI cannot deliver on its promise fast enough, it doesn't look like there's going to be an easy way out for a single stakeholder. Daybreak is produced from the newsroom of the Ken

India's first subscriber-focused business news platform. What you're listening to is just a small sample of our subscriber-only offerings. A full subscription offers daily long-form feature stories, newsletters and a whole bunch of premium podcasts. To subscribe, head to the Ken.com and click on the red subscribe button on the top of the Ken website. Today's episode was hosted and produced by my colleague,

Rachel Vargis, and edited by the Ken. Rajiv Sien.