Let me save you some time.

Yes.

Okay, not quite that simple but closer to yes than most people are comfortable admitting. And if you're reading this while scrolling through your work emails, half-distracted, quietly hoping this is one of those "AI is overhyped" reassurance pieces, I'm sorry. It's not.

I'm a senior program manager at one of the world's largest companies. I've watched AI reshape how work gets done from the inside. I've used it, fought it, been impressed by it, been annoyed by it, and if I'm honest, been quietly unsettled by it. What I'm about to tell you isn't speculation. It's what I see happening in real organizations, in real time.

Let's get into it.

Before we talk about whether AI will replace you, let's talk about what's already happened.

In 2025 alone, AI was explicitly cited as a cause of nearly 55,000 U.S. layoffs, according to consulting firm Challenger, Gray & Christmas. That's not "restructuring." That's not "strategic realignment." That's companies saying, out loud, that a machine now does what a human used to do.

Against that backdrop, the broader layoff numbers are staggering. Globally, nearly 245,000 tech jobs were cut in 2025. In 2026, the pace has accelerated with nearly 60,000 tech jobs gone in just the first three months, an average of 704 jobs lost per day, according to layoff tracker TrueUp. Not all of these were caused by AI. Companies restructure for many reasons: cost pressures, strategic pivots, leadership changes, macroeconomic conditions. But AI sits in the background of almost all of them. Either as the stated reason, the quiet accelerant, or the justification for not backfilling the roles that disappear.

Here's what the companies themselves said:

Block, the company behind Square and Cash App, is the clearest case. CEO Jack Dorsey cut 4,000 employees in early 2026, nearly 40% of the entire workforce, and was unusually direct about why: "This is not driven by financial difficulty, but by the growing capability of AI tools to perform a wider range of tasks." The day he announced it, Block's stock rose 20%. The market rewarded him for it.

Salesforce confirmed that 4,000 customer support workers were cut with AI handling the support function instead. CEO Marc Benioff said so explicitly.

Microsoft eliminated approximately 15,000 jobs through 2025. CEO Satya Nadella framed it around AI, writing to employees about "reimagining our mission for a new era" and building an "intelligence engine" rather than a software factory. Microsoft didn't say AI caused all of those cuts. But they made clear AI is reshaping what they need humans for.

Amazon cut 14,000 corporate roles in October 2025, then came back in January 2026 and cut 16,000 more for a total of 30,000 roles cut in its largest layoff in history. Their stated reasons centered on flattening management layers and moving faster. AI investment was the parallel headline: $80 billion in AI capital expenditure in 2025 alone. Revenue that year: a record $716.9 billion. Amazon is not cutting because they're struggling. They're cutting while growing, and rebuilding leaner with AI in the gaps.

TikTok conducted three separate rounds of layoffs in its U.S. e-commerce unit in a single year, without publicly attributing them to any single cause.

The honest read across all of this: AI is not the villain in every layoff announcement. But it is the force that keeps shrinking the number of humans needed to produce the same or greater output. Companies don't always say "AI did this." They don't have to. The math speaks for itself.

And here's the part that should make you put your phone down: the companies cutting are posting record profits while doing it. This isn't cost-cutting from weakness. It's optimization from strength. That distinction matters enormously for what comes next.

Let's be specific. If you work in project management, program management, or product management, here's what AI already does better, faster, or cheaper than a human.

Documentation: The static Product Requirements Document (PRD) is dying. Kalyan Ganapathisubramanian, a Group PM at YouTube with over 11 years of experience across YouTube and Google, put it bluntly: AI can now replace months of research with high-fidelity prototypes in hours. The era of the PM who creates value by writing detailed specs is ending. AI synthesizes research, generates requirements, and produces documentation at a speed no human can match.

Summarization: Got a 60-page program brief? A two-hour meeting recording? A 300-message Slack thread? AI reads it, summarizes it, and extracts action items in minutes. This used to be a PM's core skill — managing information. Now it's a prompt.

Status reporting: I know, because I do it myself: paste raw notes into ChatGPT, ask for a structured executive update, review and refine. What used to take 45 minutes takes 10. If AI saves every PM 35 minutes per status report and you write four a week, that's over two hours of PM time reclaimed, per person, per week. Multiply that by a team of 20 PMs and a CFO starts asking questions.

Risk analysis. Pattern recognition across large datasets is something AI does extraordinarily well. Identifying which projects are at risk based on velocity, communication patterns, and historical data? That's increasingly a machine job.

Scheduling, resource allocation, dependency mapping. Tools like Microsoft Copilot and enterprise AI platforms are already doing this inside organizations. The grunt work of program management is being automated layer by layer.

Entry-level coordination. The junior PM who sets up meetings, tracks action items, and updates project trackers? That role is in serious trouble. Not next year. Now.

Not all PM roles are equally exposed. Here's an honest breakdown:

High risk — significant portions already automatable:

Medium risk — humans still needed but fewer of them:

Lower risk — for now:

Notice where the lower risk category lives: at the top of the seniority ladder, and in roles where the work is fundamentally human involving politics, trust, relationships, judgment calls that require context no AI currently has.

That's not an accident. But it's not a guarantee either.

Here's the part of my job I always assumed was AI-proof: being on calls, managing relationships, reading the room, navigating the politics of a complex program across multiple teams and executives.

Can AI replace that?

Honestly? Not yet. But the question isn't whether AI can replace it today. The question is how much of everything around it gets automated, leaving a smaller, more senior, more human-judgment-dependent role.

Think of it this way. If AI handles documentation, summarization, scheduling, reporting, and analysis, then the coordination layer that currently justifies having five PMs on a program might only need two. Those two will spend 100% of their time on the things AI can't do: the calls, the relationships, the judgment. They will be more valuable, more senior, and more expensive. And there will be fewer of them.

That's not replacement. That's compression. And if you're not in the top tier when the compression happens, it doesn't matter that your role wasn't fully automated. The org chart just got smaller.

Here's where I land after seeing this from the inside:

AI will not replace humans. But humans who use AI will replace humans who don't. And organizations will use AI to justify needing fewer humans overall.

The PMs who survive this, and even thrive, are not the ones who can write the best BRD or PRD. They're the ones who can orchestrate AI to produce better outcomes than a team of humans could before.

The jobs that disappear are the ones where the primary value was execution, that is, doing the thing. The jobs that remain, and grow, are the ones where the value is deciding what thing to do and ensuring it lands.

That's a different skill set. It's learnable. But it requires you to start now.

You don't need a career overhaul this week. You need one small move.

Pick one part of your job, just one, and figure out how to do it with AI. Not instead of doing it yourself. Alongside. Learn what it can do, where it fails, and where your judgment adds something the AI can't. That's the foundation of being someone who uses AI rather than someone AI replaces.

If you want the most practical, honest guide to working alongside AI right now, Wharton professor Ethan Mollick's Co-Intelligence: Living and Working with AI is the one book I'd recommend starting with.

The people who will be fine are not the ones who are the smartest or the most senior. They're the ones who are the most curious, the most adaptive, and the most willing to learn in public.

That's the game now.

AI already has your job, or at least a meaningful portion of it. The 245,000 tech jobs cut in 2025 are not an anomaly. They are a preview. The question is not whether your industry will be affected. It's when, and whether you'll be positioned on the right side of it when it happens.

The goal of this article was not to scare you. It's to wake you up, because the people who will be fine are not the ones who buried their heads and hoped for the best. They're the ones who saw it coming, moved early, and positioned themselves for what comes next.

Speaking of which, next week, we're talking about exactly that. What to do about it. How to position yourself. How to make AI work for your career instead of against it.

Stay curious. Stay sharp. And don't sleep on this one. 

If this resonated, subscribe to the newsletter. Every week I write about AI, work, and building a career that survives whatever comes next — from someone inside one of the world's largest companies who is figuring it out alongside you.

Sources: Challenger, Gray & Christmas · layoffs.fyi · InformationWeek · IBTimes UK · CNBC · Visual Capitalist · Kalyan Ganapathisubramanian, Group PM at YouTube — AI Product Leadership session

Disclosure: This post contains affiliate links. If you click and purchase I may earn a small commission at no extra cost to you.