Something happened in February that I think deserves more attention than it got. Sam Altman, the CEO of OpenAI, publicly agreed that companies are “AI washing” their layoffs. His exact point was that some companies are blaming AI for cuts that are really driven by financial pressure, restructuring, or plain old cost reduction. The CEO of the company that arguably started the current AI wave said, on the record, that businesses are using his technology as a scapegoat. A few weeks later, Bloomberg published a piece titled “The AI-Washing of Job Cuts Is Corrosive and Confusing.” A Harvard Business Review study found that 60 percent of executives who made headcount reductions did so in anticipation of AI efficiencies, not because AI had actually replaced anyone. Only 2 percent said they made large staff reductions as a result of actual AI implementation. Two percent. That is the real number. Everything else is a story companies are telling to make cuts look strategic instead of what they usually are: cost reduction dressed up in futuristic language. I have been watching this from two angles. As someone who writes about AI coding tools and agentic workflows, I know what AI can and cannot do in a development context. And as someone who has been through hundreds of interviews and experienced the strange reality of hiring right now, I have seen the disconnect between what companies say and what they actually do. AI coding tools agentic workflows hundreds of interviews strange reality of hiring right now The data tells a clear story. Let me walk through it. What AI Washing Actually Looks Like AI washing is when a company uses AI as a justification for decisions that were going to happen regardless. The term borrows from “greenwashing,” where companies claim environmental responsibility while doing nothing meaningful. In the layoff context, it works like this: a company needs to cut costs. Instead of saying “we overhired during the zero-interest-rate era and now we are correcting,” they say “we are leveraging AI to improve efficiency and restructuring our workforce accordingly.” The first statement sounds like a failure. The second sounds like innovation. The incentives are obvious. Stock prices respond better to “investing in AI” than to “we made bad hiring decisions.” Remaining employees feel less threatened when cuts are framed as strategic modernization rather than financial panic. And the executives making these decisions look forward-thinking instead of reactive. Here are some of the numbers from 2026 so far: Over 45,000 tech workers have been laid off in the first three months of 2026
More than 9,200 of those cuts were explicitly attributed to AI and automation
A survey of hiring managers found that nearly 60 percent plan to conduct layoffs this year, with AI or automation as the most-cited reason
But only 9 percent said AI has fully replaced certain roles
An NBER study surveying almost 6,000 executives across the US, UK, Germany, and Australia found that over 80 percent of firms report no impact from AI on either employment or productivity over the past three years Over 45,000 tech workers have been laid off in the first three months of 2026 More than 9,200 of those cuts were explicitly attributed to AI and automation A survey of hiring managers found that nearly 60 percent plan to conduct layoffs this year, with AI or automation as the most-cited reason But only 9 percent said AI has fully replaced certain roles An NBER study surveying almost 6,000 executives across the US, UK, Germany, and Australia found that over 80 percent of firms report no impact from AI on either employment or productivity over the past three years Read that last point again. Eighty percent of firms report no impact. Zero. Nothing changed. Yet, AI is the number one reason companies give for cutting headcount. That gap between narrative and reality is what AI washing is. Why Companies Do This I do not think most executives are sitting in a room consciously plotting to deceive people. The dynamic is more subtle than that. There is genuine uncertainty about what AI will do to jobs. Nobody knows exactly how fast AI capabilities will advance or which roles will be affected first. In that uncertainty, it is rational (from a corporate strategy perspective) to preemptively reduce headcount in areas you think might become automatable. You do not want to be the company that waits too long. The problem is that “preemptively reducing headcount based on speculation” is not the same thing as “AI replaced these jobs.” One is a bet. The other is a fact. Companies are presenting the first as the second. The Harvard Business Review study nailed this distinction. Companies are laying off workers because of AI’s potential, not its performance. They are making cuts based on what AI might do in two to three years, then telling the market it is already happening. This matters because the narrative shapes reality. When every company says AI is replacing workers, other companies feel pressure to do the same thing. Investors start expecting it. Job seekers start believing it. A story that started as corporate positioning becomes a self-fulfilling prophecy that distorts the entire labor market. The Junior Developer Crisis Is Real and Getting Worse While the AI washing conversation is about spin, the junior developer crisis is about something much more concrete. And the two are deeply connected. The numbers are genuinely alarming: Entry-level developer hiring has dropped 73 percent in the past year
In the UK, entry-level tech roles fell 46 percent in 2024, with projections hitting 53 percent by the end of 2026
CS graduate unemployment is at 6 to 7 percent, up from historical lows
Tech internship postings have declined 30 percent since 2023
The average tech job search now takes five to six months and requires 200+ applications Entry-level developer hiring has dropped 73 percent in the past year In the UK, entry-level tech roles fell 46 percent in 2024, with projections hitting 53 percent by the end of 2026 CS graduate unemployment is at 6 to 7 percent, up from historical lows Tech internship postings have declined 30 percent since 2023 The average tech job search now takes five to six months and requires 200+ applications The “Magnificent Seven” (Alphabet, Amazon, Apple, Meta, Microsoft, NVIDIA, Tesla) have sharply reduced campus hiring. When these companies cut junior hiring, smaller companies follow. The entire pipeline contracts. The stated reason is always the same: AI can do what juniors used to do. Why hire someone to write boilerplate, unit tests, or simple CRUD endpoints when an AI agent handles it? And here is where AI washing intersects with the junior developer problem. Some of this is real. AI tools genuinely do handle tasks that used to go to junior developers. But the 73 percent decline is not explained by AI capability alone. It is amplified by the AI washing effect, where companies use the AI narrative to justify not investing in talent development that they were already reluctant to fund. Training junior developers has always been expensive. It takes 12 to 18 months before a junior contributor is net positive. In a high-interest-rate environment where every hire needs to deliver ROI from day one, junior investment was already under pressure. AI gave companies a socially acceptable reason to stop doing it entirely. And for the juniors who do make it through, the technical interview process itself is shifting in ways that add another layer of difficulty. technical interview process itself is shifting The Pipeline Problem Nobody Wants to Talk About Here is what concerns me about this, beyond the immediate hardship for people trying to break into the industry. If you stop hiring juniors for three to five years, you do not just have a “junior developer shortage.” You have a mid-level developer shortage three years later. And a senior developer shortage five years after that. The pipeline is cumulative. Every year of reduced junior hiring compounds into bigger gaps down the line. Senior developers do not appear from nowhere. They are juniors who got opportunities, made mistakes, learned from production systems, and accumulated experience over the years. There is no shortcut for this. AI can accelerate learning, but it cannot replace the experience of shipping real software in a team, dealing with production incidents, navigating technical debt, and learning to make architectural decisions under uncertainty. The industry is acting as if the supply of experienced developers is infinite. It is not. And the companies that are cutting junior hiring most aggressively today will be the ones complaining loudest about talent shortages in 2029. Forrester’s 2026 predictions actually hint at this: they forecast that over half of layoffs attributed to AI will be quietly reversed as companies realize the operational challenges of replacing human talent prematurely. The correction is coming, but the damage to the junior pipeline is already done. What AI Actually Replaced (And What It Did Not) I use AI coding tools daily. I wrote about how I use them and about the shift to agentic coding. I am not someone who dismisses what these tools can do. how I use them the shift to agentic coding Here is my honest assessment of what AI has actually replaced in a development context as of March 2026: Things AI handles well enough to reduce headcount: Things AI handles well enough to reduce headcount: Simple boilerplate generation (scaffolding, CRUD, basic API endpoints)
Writing unit tests for existing code
Code documentation
Basic bug fixes and error resolution
Repetitive refactoring tasks Simple boilerplate generation (scaffolding, CRUD, basic API endpoints) Writing unit tests for existing code Code documentation Basic bug fixes and error resolution Repetitive refactoring tasks Things AI does not replace, despite the marketing: Things AI does not replace, despite the marketing: Understanding a business domain deeply enough to make good product decisions
Debugging complex, multi-service production issues where the problem is in the interaction between systems
Making architectural decisions that account for team size, operational capacity, and business constraints
Code review that catches not just bugs but design problems
Mentoring and knowledge transfer between team members
Building relationships with stakeholders to understand what they actually need versus what they say they want Understanding a business domain deeply enough to make good product decisions Debugging complex, multi-service production issues where the problem is in the interaction between systems Making architectural decisions that account for team size, operational capacity, and business constraints Code review that catches not just bugs but design problems Mentoring and knowledge transfer between team members Building relationships with stakeholders to understand what they actually need versus what they say they want The first list is real. Those tasks genuinely take less human time than they used to. But they were never the full scope of what a developer does. They were the mechanical parts, the parts that were already the easiest to commoditize before AI existed. The second list is where actual developer value lives. And none of it is going anywhere soon. The gap between what AI can do and what companies claim AI can do is the AI washing problem in a nutshell. The technology is genuinely useful. The claims about it are wildly exaggerated. And the exaggeration is causing real harm to real people. What This Means If You Are a Developer Right Now If you are mid-career or senior, AI washing is mostly noise. Your skills are in demand, the talent shortage at the experienced level is real, and the companies doing performative layoffs are not the ones you want to work for anyway. The practical impact on your career is manageable. If you are a junior or trying to break in, the situation is genuinely difficult, and I am not going to pretend otherwise. The entry points are shrinking. The competition for remaining junior roles is intense. And the “just learn AI” advice that gets thrown around constantly is not a complete answer. Here is what I actually think matters right now: Build things that cannot be vibe-coded. I wrote about why I avoid vibe coding for real products. The principle applies to your portfolio, too. If an AI can generate your project in five minutes, it does not demonstrate anything. Build something that requires understanding a real problem, making trade-offs, and thinking about edge cases. That is the signal that separates you from a prompt. Build things that cannot be vibe-coded. why I avoid vibe coding Focus on the skills AI cannot replicate. System design, debugging complex distributed systems, understanding performance bottlenecks, communicating technical concepts clearly, and working effectively in a team. These are the skills that make a developer valuable at any level, and they are exactly the skills that AI cannot handle. Focus on the skills AI cannot replicate. Do not let the AI washing narrative define your self-worth. If you are struggling to find a job, it is not because you are obsolete. It is because the market is broken in specific, identifiable ways that have more to do with corporate finance than with technology. The companies that are not hiring juniors right now are making a short-sighted decision that they will regret. That does not help you today, but it is the truth. Do not let the AI washing narrative define your self-worth. Consider the indie path. I have been building and launching side projects for over a year now. The barrier to entry for building and shipping your own product has never been lower. If the traditional hiring pipeline is broken, building something yourself is a legitimate alternative path to building credibility, income, and skills simultaneously. Consider the indie path. side projects What This Means If You Run a Company If you are a founder or engineering leader, I want to make one direct point. The juniors you do not hire today are the seniors you cannot hire in five years. Every company that cuts junior hiring is free-riding on the companies that still invest in developing talent. If everyone free-rides, the supply runs out. I understand the economics. I understand that training costs money and that investors want efficiency. But efficiency without pipeline investment is just borrowing from the future. The companies that will have the strongest engineering organizations in 2028 and beyond are the ones that kept hiring and developing juniors through this period. They will have the institutional knowledge, the team depth, and the leadership bench that you cannot build overnight or buy on the open market. And for what it is worth, the AI washing angle is a reputational risk. When Forrester says over half of AI-attributed layoffs will be quietly reversed, that tells you something about how sustainable the current approach is. The market will remember which companies used AI as a cover story. The Honest Assessment AI is a powerful technology that is genuinely changing how software gets built. I use it every day, and I am more productive because of it, though recent research on AI brain fry shows that even productivity gains come with cognitive costs the industry is not talking about. That is not in question. AI brain fry What is in question is the gap between what AI actually does and what companies claim it does, and the real human cost of that gap. When you inflate AI’s capabilities to justify layoffs you were going to make anyway, you distort the labor market, damage trust, and undermine the pipeline that produces the talent you will need in the future. AI washing is not just dishonest. It is strategically stupid. And the junior developer crisis it is helping accelerate is a problem the entire industry will pay for, not just the people being shut out of opportunities today. The technology is real. The hype is manufactured. Knowing the difference matters more right now than it ever has.

This story contains new, firsthand information uncovered by the writer.

Microsoft

Amazon

NVIDIA

The Rules Have Changed But Nobody Was Informed: Navigating Technical Interviews in 2026

What Is AI Washing? Layoff Lies and the Junior Developer Crisis Nobody Is Doing Anything About

About Author

Comments

TOPICS

THIS ARTICLE WAS FEATURED IN

Related Stories

Building the "Spotify Wrapped" of Product Hunt: A 48-Hour Technical Sprint to Map the Maker Economy

10 Google Artificial Intelligence Tools Available to Everyone

174 AI Tools To Try

5 Artificial Intelligence Tools to Simplify Your Life: No Cost Required

6 AI Tools Software Devs Should Have On Their Radar

6 Best-In-Class AI Tools Every CTO Needs to Know

Building the "Spotify Wrapped" of Product Hunt: A 48-Hour Technical Sprint to Map the Maker Economy

10 Google Artificial Intelligence Tools Available to Everyone

174 AI Tools To Try

5 Artificial Intelligence Tools to Simplify Your Life: No Cost Required

6 AI Tools Software Devs Should Have On Their Radar

6 Best-In-Class AI Tools Every CTO Needs to Know

Light-Mode

Classic

Newspaper

Minty

Dark-Mode

Neon Noir

Minty

HN StartUps