Fast Facts
A former Block machine learning engineer named Kenji helped build the AI tools that ended up making his role redundant. His story reflects a pattern now visible across 30,000+ tech layoffs in 2026 — where workers actively trained the systems that displaced them. This piece breaks down why it’s happening, what the data actually says, and the specific moves every technical worker needs to make right now — regardless of industry or geography.
When the Tools You Built Come for Your Job
The phrase “machine learning engineer AI layoffs” would have sounded almost absurd two years ago. These were the builders — the people who understood the stack, trained the models, and sat closest to the technology. If anyone was protected, surely it was them.
Kenji thought the same. A former machine learning engineer at Block, the fintech company led by Jack Dorsey, he spent the last year of his employment doing exactly what he was instructed to do: integrating AI tools across his workflows, streamlining repetitive tasks, and handing over predictable processes to automated systems. He was efficient. He was compliant. And then he was gone.
According to Futurism’s reporting, Kenji described the experience with a clarity that is hard to dismiss: “At some point you look around and say, ‘Gosh, I’m not doing that much of the work anymore, am I?'”
His story is not an isolated case. It is the sharpest, most human illustration of a structural shift that is quietly reordering technical employment — in tech, finance, logistics, manufacturing, and increasingly everywhere that data-driven workflows exist. The central question this piece addresses is not whether AI is replacing workers. That debate is already behind us. The real question is: why are the most capable technical workers often the first to go — and what does that mean for everyone else?
Truth 1
Why Machine Learning Engineer AI Layoffs Are Happening to the Workers Who Built the Tools
Kenji’s account reveals a mechanism worth understanding carefully. When Block pushed its engineering teams to adopt AI tools, those engineers — being technically fluent — documented their processes, refined prompts, and transferred repeatable tasks to automated systems. That transfer became, in effect, a capability audit.
“Over the last year that we were strongly encouraged to use all these AI tools, we were laying the foundations for our own replacement. If you show the tool how to do a task once or twice, it can kind of take it from there.”— Kenji, former Block ML engineer, via Futurism
Block CEO Jack Dorsey made his reasoning explicit in a shareholder letter, stating that “intelligence tools have changed what it means to build and run a company.” The company cut nearly 4,000 roles — almost half its workforce — in February 2026, according to Programs.com’s company-by-company breakdown of AI layoffs.
The sequence follows a recognizable pattern: adopt mandate arrives first, the workflow documentation follows, the redundancy assessment comes next, and the headcount reduction is last. Workers are not being eliminated because they failed. They are being eliminated because they succeeded — at teaching the machine to do their job.
Truth 2
Why Over 30,000 Technical Workers Have Already Lost Jobs to AI in 2026 — And the Real Number Is Likely Higher
30,000+
Tech employees impacted by AI-driven layoffs in 2026 alone, with 45+ CEOs explicitly citing AI efficiency as the driving reason.
Source: Programs.com
The scale becomes clearer when you look at the company-level data. Amazon cut 30,000 roles across late 2025 and early 2026. Oracle is reportedly eliminating close to the same number as it reallocates resources toward AI infrastructure. Baker McKenzie, a global law firm, cut between 600 and 1,000 support roles — research, marketing, secretarial functions — citing AI-driven efficiency reviews.
Critically, these cuts are no longer confined to Silicon Valley. HR Executive’s analysis of Forrester’s Predictions 2026: The Future of Work report found that AI-driven layoff announcements are now coming from finance, logistics, consulting, media, retail, and manufacturing — sectors that were not part of the conversation 18 months ago.
For operations analysts, maintenance engineers, and process designers working in automated manufacturing environments, the trajectory is becoming harder to dismiss. Our earlier analysis on AI washing explored how companies use automation as narrative cover — but the underlying structural pressure on execution-layer roles is accelerating regardless of how companies frame it publicly.
Truth 3
Why Companies Are Laying Off Workers for AI That Doesn’t Exist Yet — And Who Pays for That Bet
Harvard Business Review published a finding in February 2026 that reframes the entire conversation. According to their analysis, many companies are cutting headcount based on AI’s potential rather than its current, proven performance. The workforce absorbs the cost of that bet in real time, while the technology catches up at its own pace.
The consequences of that bet failing are well-documented. Klarna replaced 700 employees with AI agents. Service quality declined. Customer satisfaction fell. The company was forced to course-correct. According to Forrester’s research cited by HR Executive, 55% of employers report regretting AI-driven layoffs — yet most are not reversing them at equivalent salary levels.
55%
of employers report regretting AI-driven layoffs. Most quietly rehire offshore at lower salaries rather than restoring original roles.
Source: Forrester Predictions 2026 via HR Executive
Forrester’s prediction is specific: half of AI-attributed layoffs will result in quiet rehiring — but offshore and at significantly lower pay. The AI announcement provides the narrative. The margin improvement does the actual work. Workers in emerging markets who have historically benefited from tech outsourcing face a compounded pressure: fewer outsourceable tasks, and lower rates on those that remain.
Truth 4
Why Entry-Level Technical Roles Are Disappearing Faster Than Any Other Category in 2026
Anthropic’s own labor market research — published via Fortune — found a 16% fall in employment in AI-exposed roles among workers aged 22 to 25. Entry-level positions are being eliminated faster than senior roles — which means the pipeline feeding technical expertise into industry is narrowing at exactly the moment AI literacy is most in demand.
Forrester’s data adds a specific dimension to this. Only 16% of individual workers had high AI readiness (what Forrester calls AIQ) in 2025. That number is projected to reach just 25% in 2026. Only 23% of AI decision-makers offered prompt engineering training to employees in 2025. Workers are largely self-teaching.
The paradox here is direct: Gen Z workers have the highest AIQ at 22%, compared to just 6% for Baby Boomers — yet companies are disproportionately cutting entry-level positions, eliminating the cohort most prepared to work alongside these systems.
⚠ Fiction — Illustrative Scenario
Amara spent two years building her firm’s internal data labeling pipeline. When the company introduced a foundation model that could handle 80% of her annotation tasks in March 2025, she trained it — refining edge cases, correcting outputs, improving its confidence thresholds. Eight months later, her contract was not renewed. The model her work improved is now handling what her team of five once managed. She found out via an automated HR email at 7 a.m. on a Tuesday.
Truth 5
Why Upskilling Alone Will Not Protect Technical Workers — And What Actually Will
Kenji’s final observation to Business Insider deserves to be taken seriously: “If I land a job tomorrow, I have zero confidence that it, too, couldn’t be automated away in a couple years.”
That is not defeatism. It is an accurate reading of current conditions. And the response it demands is not another certification or another tool to learn — it is a repositioning around the categories of work that automated systems consistently struggle with: judgment under genuine ambiguity, governance of systems that fail in unexpected ways, stakeholder negotiation across competing interests, and accountability for decisions that cannot be rolled back.
Shawn K, a software engineer with two decades of experience who lost his $150,000-a-year job to AI displacement, described it to Fortune after sending 800 applications: “Companies are using AI to save money by cutting talent — rather than leveraging its power and embracing cyborg workers.” The workers most at risk are those whose value can be described entirely by their outputs — without reference to the judgment, context, or accountability required to produce them.
Roger Lee, founder of Layoffs.fyi, told AOL News: “I do think 6 years of persistent layoffs have led many tech workers to re-evaluate the perceived ‘safety’ of tech jobs and their relationship with the industry overall.” According to that same report, there have been more than 35,000 layoffs in the tech sector worldwide so far in 2026, with close to half attributable to Amazon alone.
The workers building durability into their careers right now are not those accumulating tools — they are those positioning themselves as the people who decide how the tools are governed, audited, and corrected when they fail. That is the role automation cannot easily absorb. Our analysis on stakeholder fear and AI retraining budgets captures exactly why organizations consistently underinvest in developing this capacity internally — creating a durable gap that informed workers can occupy.
Global Context
What the AI Layoff Pattern Means for Technical Workers Across Every Market
The US Bureau of Labor Statistics reported that employers shed 92,000 jobs in February 2026, with the unemployment rate ticking up to 4.4%, according to Fortune. CNN’s labor market analysis noted the unemployment rate in January sat at 4.3% — half a percentage point higher than late 2023, when the generative AI investment cycle began. These numbers do not confirm economic collapse, but they confirm a directional shift.
For technical workers in Asia, Africa, Latin America, and Eastern Europe — markets where tech outsourcing has historically built middle-class employment — the Forrester offshore rehiring prediction represents a structural compression on both sides: fewer outsourceable tasks available, and lower rates on those that survive. The calculation that made offshore technical work economically attractive is being disrupted from two directions simultaneously.
Nigeria’s growing AI pilot project ecosystem illustrates both the opportunity and the alignment challenge this creates. The roles that survive and grow in this environment are not those executing known processes at scale — they are those designing, auditing, and governing the systems doing the execution. That repositioning is available to any technically literate worker willing to pursue it deliberately rather than reactively.
For anyone tracking where automation investment is flowing and where workforce pressure will intensify next, this piece on AI resentment building inside industrial workplaces captures the human dimension of what these numbers represent at the operational level.
Further Reading & Related Analysis
- The Truth Behind AI Layoffs — Or AI Washing?
- AI Resentment Is Rising Inside Industrial Workplaces
- How Stakeholder Fear Kills AI Retraining Budgets Mid-Cycle
- Industrial AI Pilot Projects in Nigeria: What’s Actually Working
- The AI Productivity Paradox: Why Automation Promises More Than It Delivers
Frequently Asked Questions
Why are machine learning engineers being laid off if AI still needs human oversight?
Because the specific tasks ML engineers perform — data labeling, pipeline optimization, model fine-tuning — are precisely the tasks that current AI tools are fastest at absorbing. Human oversight remains necessary at the governance and architecture level, but execution-layer roles are disappearing as those tasks become automatable. Engineers who made their work legible to AI systems inadvertently made their positions redundant.
How many jobs has AI eliminated in the tech sector in 2026?
According to Programs.com, over 30,000 tech employees have been impacted by AI-attributed layoffs in 2026 alone, with 45+ CEOs explicitly citing AI efficiency. Layoffs.fyi puts the broader tech sector figure at 35,000+. The US Bureau of Labor Statistics reported 92,000 total job losses across all sectors in February 2026.
Is AI actually responsible for layoffs or are companies using it as an excuse?
Both dynamics are real. Harvard Business Review documented that many companies are cutting roles based on AI’s potential rather than current performance — betting on future capability while reducing immediate payroll. But structural automation pressure on pattern-recognition and workflow-execution roles is genuine, regardless of how layoffs are communicated publicly. The financial motive and the technological motive often run in parallel.
Will companies rehire the workers they let go for AI?
Forrester Research predicts half of AI-attributed layoffs will result in quiet rehiring — but offshore and at significantly lower salaries. 55% of employers already report regretting AI-driven cuts. Klarna’s public reversal after replacing 700 employees with AI agents is the clearest documented example of this pattern playing out.
Which technical roles are most protected from AI displacement in 2026?
Roles requiring judgment under ambiguity, governance of AI systems, stakeholder negotiation, and accountability for irreversible decisions hold the strongest position. Workers who understand where AI fails — not just how to use it — are significantly harder to displace than those whose value is measured purely by outputs that can be replicated.
How does the AI layoff trend affect workers in non-Western markets?
Forrester’s offshore rehiring prediction suggests that as Western companies cut senior technical roles, some of that work will migrate to lower-cost markets — but at compressed rates. Workers in Asia, Africa, and Latin America face a dual pressure: fewer outsourceable tasks overall, and lower pay on those that remain. The career durability advantage increasingly lies in governance and system-design roles, not execution roles.
Don’t Get Caught Off Guard by the Next Shift
CreedTec breaks down what automation decisions actually mean — for operators, analysts, and technical workers navigating a market that is moving faster than the headlines suggest.
Subscribe to CreedTec
