AI accelerates: Is humanity in danger?

Between Planned Obsolescence of Human Labor and the Search for a New Social Contract

We are living through a pivotal moment in human history. Artificial intelligence is no longer content with automating repetitive tasks: it is now striking at the very heart of intellectual work, profoundly reshaping our relationship with employment, value, and the meaning of existence.

Recent figures reveal a brutal acceleration that forces us to ask a question as vertiginous as it is urgent:

Is humanity programming its own economic obsolescence?

The Layoff Tsunami: When Numbers Speak for Themselves

Key Data – January 2026

600,000 jobs lost worldwide in January 2026, a 15% increase over the previous year.
More than 180,000 layoffs in the tech sector alone in 2025.
110,000 layoffs in the United States for just 5,000 job creations, an alarming ratio of 1 to 20.

Early 2026 data confirm what many have long suspected: we are no longer dealing with a mere cyclical adjustment but a true structural transformation of the labor market.

In January 2026, nearly 600,000 jobs were cut worldwide, marking a 15% acceleration compared to the previous year.

In the U.S. tech sector, 76,214 positions have been eliminated since January 2025—a 35% increase compared to 2024.

What is particularly striking is the paradoxical nature of this wave of layoffs.

The companies carrying out massive job cuts are the very ones investing billions in AI.

Amazon announced 16,000 job cuts in January 2026, even as it ramped up investments in artificial intelligence development.
Microsoft laid off more than 15,000 employees in 2025, justifying the cuts by the growing efficiency of its internal AI: 30% of its code is now generated by GitHub Copilot, mechanically reducing the need for junior engineers.

The banking sector is no exception: Crédit Commercial de France has cut 1,400 out of 2,400 positions. Société Générale plans 5,000 layoffs by 2026.

BNP Paribas is moving toward a gradual reduction of 1,000 to 1,400 jobs annually.
Intel plans to cut 24,000 jobs, roughly 22% of its global workforce. Meta has let go of 1,500 employees from its Reality Labs division.
Tata Consultancy Services eliminated 12,000 jobs in the third quarter of 2025. The list grows longer each month, now affecting all sectors: from tech to banking, administration to services.

Beyond the Numbers: Who Are the Victims?

Contrary to what one might expect, it’s not only low-skilled jobs that are at risk. The data reveal that knowledge-based professions are at the forefront.

Software engineers, developers, HR professionals, financial analysts, project managers—all professions once thought safe are now taking the full brunt of generative AI tools.

The Indeed report shows that AI primarily impacts entry-level jobs across all sectors, notably marketing, administrative assistance, and human resources.

Job offers targeting workers with at least five years of experience fare better, creating a new generational divide.

For the first time since the 1980s, the professional integration of new graduates is deteriorating.

The Obsolescence of Biology

The current revolution fundamentally differs from the Industrial Revolution. Back then, machines amplified human physical strength; today, AI replaces the capacity for oversight, analysis, and orchestration—once the final stronghold of human labor.

A Multidimensional Cognitive Superiority

Today’s AI models are no longer just assistive tools. They can code in 30 different languages, summarize 1,000 research articles in 15 minutes, and interpolate results with precision that sometimes surpasses humans in specific tasks.

They represent the aggregation of all indexable knowledge ever produced by humanity, accessible and usable in seconds.

At Microsoft, 30% of code is now generated by GitHub Copilot. At IBM, the AskHR system has replaced 8,000 human resources staff, saving $3.5 billion.

These figures are just the start of a much broader trend: according to the World Economic Forum, 41% of global employers explicitly plan AI-related workforce reductions in the next five years.

The Challenge of Cognitive Atrophy

Beyond job losses, the omnipresence of AI raises an existential question: what happens as we gradually delegate our cognitive abilities to machines?

According to a survey by Elon University in April 2025, 61% of 300 experts believe AI will change how we think, reason, and understand the world.

Researchers now speak of “cognitive atrophy,” a phenomenon where excessive dependence on AI leads to the decline of our analytical, memory, and critical thinking skills.

Our average attention span fell from 2 minutes 30 seconds in 2000 to just 47 seconds in 2024. As philosopher Gaspard Koenig notes, delegating our reasoning and decision-making abilities “will soon pose a real cognitive problem, with the risk of atrophy in the brain areas dedicated to decision-making.”

The Limits of Biological Speed

The human brain operates on a biological substrate roughly 1,000 times slower than electricity flowing through a wire.

In a competitive and efficiency-driven market economy, AI’s comparative advantage becomes absolute: it produces more, better, faster, and at a fraction of the cost.

This inescapable biological reality presents humanity with an unprecedented dilemma.

If humans could once fall back on orchestrating machines and overseeing complex processes, AI is now proving itself better than humans at supervising and orchestrating other AIs.

Thus, humans are losing their last objectively valuable niche in the knowledge economy.

The Collapse of the Expertise Hierarchy

The traditional work model was built on a clear hierarchy: the most repetitive and physical tasks were automated while humans retained a monopoly on supervision, creativity, and strategic decision-making.

This model is falling apart.

Current conversational AIs can now assist with strategic decision-making, generate creative ideas, analyze complex data, and even supervise other AI systems.

The MASK benchmark (Model Alignment between Statements and Knowledge), published in March 2025, does highlight AI’s current limits—none of the thirty models tested achieved more than 46% honesty—but that hasn’t slowed their mass adoption in the real economy.

The Breaking of the Social Contract

The shift toward an automated society will not happen without turbulence. Historically, periods of rapid economic transformation have always brought major social upheavals.

The question isn’t whether this transition will be difficult, but how difficult.

The Failure of the Consumption Loop

Capitalism depends on a virtuous cycle: work generates income, which fuels consumption, which drives production and creates jobs.

Mass automation breaks this cycle at its most fragile point. Without jobs, increasing numbers of people no longer have incomes.

Without income, consumption collapses. Without consumption, the capitalist system as we know it faces systemic crisis.

This implacable logic presents modern societies with a paradox: we now have the technological means to produce more with fewer workers, but our economic system is unable to redistribute the fruits of this productivity except via employment.

The Specter of Systemic Violence

Projections suggest that 30 to 50% of current knowledge-based jobs could be displaced within 10 years.

History teaches us that an unemployment rate of 20–30% is consistently correlated with major civil unrest.

The Great Depression of the 1930s, with unemployment rates over 25% in some countries, led to the collapse of democratic structures in favor of totalitarian regimes.

Sociological research shows that mass unemployment triggers a “decivilization process”: the disappearance of jobs shortens social interdependence chains, social conventions fray, and self-control diminishes.

Elites’ Response: Bunkers and Private Islands

It’s telling that numerous tech billionaires are investing heavily in underground bunkers in New Zealand or purchasing secluded islands.

These preparations show a clear anticipation: economic and tech elites expect a period of massive and rapid civil unrest.

This retreat strategy by the ultra-wealthy, rather than investing in systemic solutions, illustrates the depth of the widening divide.

While technology could in theory enable an age of abundance, we risk instead plunging into a period of unprecedented social turmoil.

Toward a New Ontology of Value: Beyond Work

If humanity manages to get through this period of transition—the notorious “eye of the needle”—it will have to invent a radically new economic system, one where human compensation is completely divorced from direct economic value.

The Paradox of Universal Basic Income

Universal Basic Income (UBI) is often proposed as a miracle solution. The idea is appealing: provide every citizen with an unconditional income sufficient to cover basic needs, thus freeing people from the constraints of wage labor and enabling personal fulfillment.

Pilot programs have shown encouraging results. In Finland, UBI recipients reported lower rates of depression, loneliness, and mental stress.

The study funded by Sam Altman in the United States, which distributed $1,000 per month for three years to 3,000 participants, showed a reduction in stress and problematic substance use.

However, UBI faces a fundamental neurological obstacle: studies suggest that when basic needs are met without effort, motivation and life satisfaction may drop, paving the way for a form of existential depression.

Our brains evolved over millions of years for action, problem-solving, and the search for meaning—not for passive idleness.

Philosopher Gaspard Koenig raises a crucial question: what will happen to people who, in their “sovereign freedom,” squander their guaranteed payments?

The stigma could become even worse. UBI alone does not resolve the questions of meaning, identity, and dignity that work has traditionally provided.

The Rise of Postmodern “Bullshit Jobs”

Just as a hunter-gatherer could never understand the value of a modern office worker, we do not yet understand what the jobs of the future will be.

These activities may no longer be tied to survival or economic production, but to fulfilling fundamental human needs: social connection, entertainment, knowledge sharing, and artistic creation.

Paradoxically, IBM illustrates this transition: after letting go of 8,000 HR employees replaced by AI, the company saw a rebound in hiring for software engineering, marketing, and AI system supervision. CEO Arvind Krishna explains:

As we automated repetitive tasks, our total headcount increased thanks to investments in high-value-added sectors.

A System Based on Human Credit

A bold alternative to UBI could be a credit system that rewards purely human values: courage, bravery, the ability to create social bonds, civic engagement, and artistic creativity.

The goal would be to preserve a sense of autonomy and meaning in a world where AI manages material production and technological evolution.

This system would rest on the acknowledgment that some activities have intrinsic value to society, even if they don’t generate direct economic profit.

Parents attentively raising their children, citizens engaged in their communities, artists creating beauty—all these contributions could be valued and rewarded in a post-work economy.

The Cognitive Challenges of Living with AI

Beyond economic and social questions, the omnipresence of AI poses existential challenges for our very cognition.

Neuroscience shows that AI rewires our cognitive circuits, shifting mental effort from complex problem-solving to mere coordination.

One study found that when we know information will be stored digitally, we make less effort to remember it—the so-called “Google effect.”

Routine use of GPS reduces our formation of spatial mental maps. Relying constantly on AI to draft emails, generate presentations, or solve problems could, over time, atrophy our core capabilities.

As highlighted in a report from Elon University, we risk a general impoverishment of human intelligence—a generation unable to sustain prolonged concentration or autonomous critical thought. These fears touch the very foundations of our cognitive dignity and mental freedom.

However, this dystopian vision is not inevitable. AI can also be a tremendous lever for intelligence and creativity—if we remain able to think, write, and create without it.

The challenge is to use AI as a tool for augmentation, not as a replacement for our faculties.

What Skills for Surviving in the Age of AI?

In the face of this transformation, some human skills are proving more resilient than others.

Data show that it’s not technical skills that protect most against automation, but core human skills.

Critical thinking and genuine creativity remain essential assets. The ability to question algorithmic output, to spot biases, to add an original perspective—these skills are hard to automate.

Emotional intelligence, the ability to connect, to show empathy—qualities that AI still cannot replicate.

The jobs most immune to AI are revealing: nurses’ aides, health workers doing blood draws, painters, embalmers.

All share a common trait: they require physical presence and direct human interaction that AI cannot reproduce.

Key Skills for 2026 and Beyond

Critical thinking: Analyzing, interpreting, and putting into perspective data produced by AI
Authentic creativity: Going beyond the predictable solutions of algorithms
Emotional intelligence: Connecting, showing empathy
Adaptability: Ability for continuous learning in a rapidly changing environment
Human–machine collaboration: Knowing how to get the best out of AI without becoming dependent on it

Scenarios for the Future: Between Dystopia and Utopia

The Dystopian Scenario: Social Fracture

In this bleak scenario, the transition to automation is chaotic and unequal. A small technological and financial elite captures most productivity gains, while a growing share of the population is economically marginalized.

Unemployment rises to 20–30%, triggering major civil unrest. Democratic structures erode amid rising populism and authoritarian regimes.

Social tensions explode, fueled by a widespread sense of uselessness and loss of dignity.

The Utopian Scenario: Post-Scarcity Society

In this optimistic scenario, humanity manages to pass through the “eye of the needle” and invent a new social contract.

The goal of civilization becomes to reach 100% unemployment, freeing humans to pursue their passions and dreams.

Productivity gains produced by AI are fairly redistributed. A system of universal income, coupled with recognition of non-market contributions, allows everyone to live in dignity while finding purpose in life.

Education, art, science, caregiving, civic engagement become the new forms of “work” valued by society.

The Most Probable Scenario: A Hybrid Path

Reality will likely fall somewhere between these two extremes. We’ll probably witness a period of significant turbulence, followed by gradual adjustments. Some countries and regions will navigate this transition better than others, creating new forms of geographic inequality.

The labor market will become fragmented: on one end, a minority of highly qualified workers able to supervise and exploit AI; on the other, a majority employed in sectors still hard to automate (care work, crafts, creativity); and in between, a disappearing middle class.

The Imperative of Redefining Meaning

The future isn’t merely an extension of the present—it’s a deep metamorphosis of our civilization.

We are heading towards a world where the goal may be to reach 100% unemployment in the traditional sense, freeing humanity for new forms of fulfillment and contribution.

Yet, to achieve this post-scarcity utopia without first falling into chaos, society must urgently find a way to redefine the meaning of human existence before instability from loss of economic purpose tears the social fabric.

The questions before us are formidable:

How can we preserve human dignity in a world where traditional work disappears?
How can we fairly redistribute the fruits of automation?
How can we maintain social bonds and a sense of usefulness?
How can we avoid the atrophy of our cognitive abilities due to excessive delegation to AI?
How can we get through the transition period without lapsing into violence and authoritarianism?

These challenges are not insurmountable, but they demand collective awareness and political will commensurate with the stakes.

History will remember this decade as the moment when humanity had to decide: to program its own obsolescence, or to reinvent the social contract for the age of artificial intelligence.

The countdown has begun.

The decisions we make today—or postpone—will determine whether our children inherit a post-work utopia or a dystopia of mass unemployment and chronic instability.

The question is no longer whether AI will radically transform our society, but how we choose to live through this transformation.

Main sources: Tech layoff data (Layoffs.fyi, 2025–2026), studies on cognitive atrophy (Elon University, 2025), UBI reports (Sam Altman/OpenResearch, Finnish studies), labor market analyses (Indeed, World Economic Forum), cognitive neuroscience research, sector reports (banking, tech, administration).