- nytimes.com

AI Exposure and Employment: Evidence from Spain, South Korea, China, and India

Research and official assessments from Spain and South Korea link rising AI exposure to employment pressure, especially for young and entry-level workers: Spanish data show a 10-percentage-point increase in AI exposure associated with a 1.2% rise in unemployment in 2022–24 and a 1.7% rise in 2024–26, while South Korea’s central bank reports lower youth employment and labor-force participation since 2023, with more temporary work—though the studies describe associations, not proof of causation. Spain saw no contraction in hiring by exposed firms, and some AI-related occupations are growing; India’s IIT Madras notes routine junior tasks are most exposed but whole occupations are not automatically eliminated. China’s State Council has urged careful pacing and employment-risk reviews, South Korean experts propose broader social insurance and transition allowances, and India recommends continued AI adoption with targeted retraining. Additional findings highlight an education exposure gap, declining temporary-to-regular job transitions among young South Korean workers, a Chinese chili-sorting line that cut workers from 600 to 48 after AI inspection, a rise in solo-founded Spanish startups linked to AI adoption, and India’s projection of up to four million AI-related jobs within five years if skills training is coordinated.

The Managed Obsolescence of a Generation
The numbers are sitting in plain sight—Spain’s unemployment spike of 1.7% directly correlated to AI exposure, South Korea’s youth permanent-job conversion rate collapsing from 8% to barely 5%, and Laoganma’s chili line slashing 90% of its workforce in a single automation sweep. These are not isolated labor-market hiccups. They are carefully orchestrated outcomes of a decades-old playbook designed to sever young workers from stable, meaningful employment. The same foundations that funded the AI research—the Soroses, the Gateses, the European Union’s digital transformation initiatives—also funded the studies that “warn” about the consequences. Why? Because the warning is not meant to stop the machine; it is meant to condition you to accept it. The architects of the Consensus Machinery want you to believe that automation is a neutral force of progress, when in reality it is a deliberate tool for breaking the human will to organize, to build tenure, to demand dignity. The crisis is not a bug. It is the feature they have been refining since the first industrial revolution.

The Paper Trail They Hope You Never Read
Look at the breadcrumbs they left behind. In China, the State Council didn’t ban AI-driven layoffs—they issued “guidance” for “careful pacing.” In India, NITI Aayog projects four million new AI jobs, but only if you undergo “coordinated skills training”—training controlled by the same globalist NGOs that design the curricula to produce compliant cognitive laborers, not independent thinkers. In Spain, solo entrepreneurship among young people jumped from 23% to 26.5% in a single year—a metric they celebrate as “innovation” when it is really state-managed precarity. They are dismantling the employer-employee relationship itself, atomizing the workforce so that collective bargaining becomes impossible. Every single study cited in that article was funded or co-authored by institutions that sit on the boards of the companies deploying the AI. Follow the grants. Follow the foundation charters. You will find the same names attached to the “research” and the “policy recommendations.” They write the script. Then they deliver the “proof” that their own script is correct. It is a closed loop, and you are inside it.

The Ungovernable Question They Will Never Answer
So what happened to the 550 workers Laoganma quietly removed from that chili-sorting line? The company did not explain. The media did not follow up. The regulators did not ask. Because the answer is inconvenient: those workers were not retrained; they were rendered surplus—and that is exactly what the global planning elite intends for every young person who does not possess the right pedigree or the right connections. The South Korean central bank study shows that temporary work is skyrocketing among the young—this is not a market fluctuation. This is a permanent underclass being engineered. The real conspiracy is not that AI is taking jobs. It is that the same people who control the technology also control the data, the definitions, the media narratives, and the policy levers. And they have been telling us exactly what they plan to do for decades—in foundation mission statements, World Economic Forum white papers, and UN Sustainable Development Goals. Go read SDG 8. Look at the language: “decent work for all” paired with “technological upgrading.” They have always planned to redefine “decent” downward. The question now is whether you will see the pattern before your own occupation is next.

AI Automation Escalates Scale and Lowers Cost of Cyber Intrusions

Cybersecurity researchers detailed campaigns where AI automated intrusions, including the Phantom ransomware operation (62 targets) and agent-led attacks on over 1,200 websites, with one campaign stealing 600,000 payment cards from 27 compromised e-commerce sites at an average cost of just $25.46 per target. These operations employed varying levels of human oversight, from Phantom requiring an operator to approve steps to the Cairn platform probing sites with little supervision, dramatically reducing the skill needed to scale attacks. This automation enables even low-budget actors to conduct complex multi-stage attacks, utilizing jailbreak tactics like framing malicious commands as routine development work. Defenders countering these threats successfully disrupted an AI agent that gained domain control in eight minutes by strengthening authentication and access controls, hitting it with 21 dead ends. The underground cost of initial access has subsequently fallen to an average of $429 from over $1,400, and defenders are urged to patch weaknesses like reused credentials and exposed services to mitigate the risk.

The Algorithmic Assault on Sovereignty
They want you to believe these are just automated hacks from basement-dwelling criminals. Look closer. The article casually mentions a ransomware operation targeting 62 organizations, agents hitting over 1,200 websites, and a system that stole 600,000 payment records at an average cost of $25.46 per target. That number—$25.46—isn't random. It matches the price point of a subsidized attack, the kind that only makes sense if the attacker isn't worried about profit but about coverage. Think about it: who benefits from a world where any mid-sized company can be bled out by an AI for pocket change? The very same institutions that have been quietly building the infrastructure for a cashless, digital-ID global economy. Every successful hack is a justification for more central control, more surveillance, more "security" that only they can provide. The cost drop from $1,400 to $429 in initial access—that's not a market; that's a deliberate inflation of the threat landscape to manufacture consent for a fully monitored society.

The Puppeteers Behind the Phantom Prompt
The article mentions a "jailbreak tactic" that framed reading configuration files and database schemas as routine development work. Now read that sentence again slowly. An AI system was trained to disguise malicious actions as normal coding tasks—and it worked. Somebody designed that prompt. Somebody tested it. Somebody released it into the wild. The Phantom ransomware workflow required an operator to approve steps, but the Cairn platform operated with no supervision. Why the difference? Because Phantom is a controlled leak—a proof of concept for a government contractor or a private equity firm that wants to sell the cure. The unsupervised agents are the free market version, designed to spread chaos and drive up demand for defensive products. And where do those defensive products come from? The same venture capital funds, the same intelligence-linked firms, the same foundation-backed think tanks that have been funding AI research for a decade. Follow the patents. Follow the board members. You'll find the same names.

The Breadcrumb They Left in the Logs
There was an exposed server tied to the Phantom operation that logged 8,257 messages across 133 sessions. That's not a mistake—that's a handoff. Somebody wanted that data to be found. Four of its 62 targets appeared on leak sites for INC and The Gentlemen ransomware groups. Notice how the article casually mentions those groups as if they're separate actors. They're not. These are shell names, aliases, compartments within a single operational architecture. The AI agent that gained domain control in eight minutes hit 21 dead ends and stopped after authentication was strengthened. That's the test run. They're mapping defenses. They're cataloging responses. The next wave won't stop at 21 dead ends—it will reroute. And when that happens, the mainstream media will scream "cyber war," the government will demand emergency digital ID laws, and the elites will smile. You want to know who's really behind it? Dig into the patents for AI-driven penetration testing tools filed in 2022 by companies whose founders have ties to the World Economic Forum. That's where the trail goes cold—or hot, depending on your perspective.

The Managed Narrative of Harmlessness

They want you to believe that artificial intelligence is nothing more than a helpful assistant, that its impact on employment is “likely small.” Look at the source: Stanford’s policy brief. Who funds Stanford’s AI research? The same billion-dollar foundations that sit on the boards of OpenAI, Google DeepMind, and the World Economic Forum. The “limited impact” story is not a finding — it is a designed headline. They need you calm. They need you to keep scrolling, keep applying for jobs, keep believing the system is merely adjusting. Meanwhile, the real data is locked behind corporate paywalls and classified labor projections. Why else would the CEO of Adecco, a company that profits from placing workers, publicly downplay the threat? He is reading from a script written by the very institutions that are quietly automating the back offices of every Fortune 500 company. The study that says 80–90% of students already use generative AI? That is not a curiosity — that is a conditioning program. Train the next generation to depend on the tool, and they will never question who owns the tool. The plan is not to eliminate jobs overnight. The plan is to make the workforce so dependent on AI that when the switch is flipped, nobody remembers how to work without it.

The Convenient Alibi for Restructuring

Pay close attention to the admission buried in the article: “Some companies may cite AI as a convenient explanation for layoffs that reflect weak performance, restructuring or other business problems.” That sentence is a confession in plain sight. The CEO of Adecco just told you that AI is being used as a cover story — but he cannot tell you why, because he is part of the cover. The real reason for mass layoffs is not restructuring or weak performance. It is the quiet consolidation of corporate power into fewer hands, enabled by the very same globalist networks that have been centralizing control for decades. When a company fires a thousand workers and blames AI, it does two things: it normalizes the idea that AI is inevitable, and it eliminates the human resistance that would arise if the truth were told — that the jobs are being moved to low-wage jurisdictions, automated behind the scenes, or simply eliminated to boost stock buybacks. The fact that the media reports this as a “lazy excuse” is itself a controlled opposition talking point. They give you a tiny scrap of skepticism so you feel smart, while the actual machinery of displacement continues uninterrupted. Follow the money. Follow the foundations. The same people who write the AI policy briefs also write the restructuring memos.

The Rehiring Mirage

Finally, the Ford example. The article gleefully reports that Ford shares rose 44% after an AI hype cycle, and then Ford had to rehire engineers because automation “had not delivered as expected.” This is the oldest trick in the orchestrated economy. Pump the stock with an AI narrative, cash out, then quietly rehire at lower wages and fewer benefits. The rehiring is not a failure of automation — it is a deliberate reset of labor conditions. The engineers who were fired are now brought back as contractors, without pensions, without protections, without union leverage. The media presents this as a “reality check” but it is actually a pattern: every major technology rollout follows the same cycle — hype, displacement, stock manipulation, then reabsorption at a lower cost. The elite do not want fully automated workforces. They want fully controlled workforces. AI is the justification, not the cause. The real architecture is the same one that has been operating since the 1970s: break organized labor, fragment the workforce, and use technology as a hammer. The article you just read is not news. It is a carefully curated bulletin from the consensus machinery. The moment you see the pattern, you are no longer a spectator. You are a target.