AI is destroying the old order and building the next one.
AI is the Fourth Turning's wildcard — a force with no historical precedent in previous crisis cycles. Unlike debt, generational dynamics, or geopolitical realignment, which follow recognizable patterns from prior eras, artificial intelligence introduces a genuinely novel variable into the equation. It is simultaneously destroying existing institutional structures and potentially building tools for whatever comes next.
The speed of capability advancement has outpaced every projection. Frontier models have reached PhD-level performance on graduate-level reasoning benchmarks, enterprise adoption has surged to 72% (up from 55% in just two years), and the impact on labor markets is already visible: entry-level knowledge work postings have declined 31% year-over-year. An estimated $487 billion in SaaS market capitalization has been destroyed as AI alternatives undercut incumbent software businesses. The disruption wave is moving from prediction to lived experience.
The crisis question is not whether AI will be transformative — it clearly will be — but who captures the value. When AI productivity gains outpace median wage growth by 12×, the economic surplus flows overwhelmingly to capital owners and early adopters rather than to the broader workforce. This dynamic compounds every other dimension of the Crisis Index: it accelerates inequality (debt lens), disrupts institutional legitimacy (social fragmentation), and reshapes geopolitical competition (power transition). AI is an accelerant poured on an already-burning fire.
What each metric measures, why it matters, and what the current reading tells us.
The destruction of SaaS enterprise value represents AI's first major structural impact on the economy. Companies built on the assumption that knowledge work required human labor — customer service, data analysis, content creation, basic coding, legal review — are being repriced as AI alternatives demonstrate comparable or superior performance at a fraction of the cost. This is not speculative: companies like Klarna have replaced hundreds of customer service agents with AI, and coding assistants are handling an increasing share of software development. The $487B figure will grow as AI capabilities improve.
Indeed's data on entry-level knowledge work postings is the labor market's canary in the coal mine. A 31% year-over-year decline in positions that traditionally served as the on-ramp to middle-class careers — junior analysts, associate consultants, entry-level developers, research assistants — signals a structural break in the labor market. If AI can perform the tasks that previously justified hiring and training junior workers, the career ladder loses its bottom rungs. The long-term implications for wage growth, skills development, and intergenerational mobility are profound.
McKinsey's 2025 survey shows that nearly three-quarters of enterprises have adopted AI in at least one business function, up from 55% in 2023. The acceleration reflects AI crossing the threshold from experimental to operational. Companies are no longer asking 'should we use AI?' but 'how fast can we deploy it?' The adoption curve suggests that the remaining 28% will face competitive pressure to adopt within 2-3 years or face existential disadvantage.
This metric captures the central distributional question of the AI era. AI-driven productivity gains are running roughly 12 times faster than median wage growth. In classical economics, productivity gains eventually flow to workers through higher wages. But the mechanism for that transmission — labor scarcity forcing employers to bid up wages — breaks down when AI can substitute for human labor at scale. The gap between productivity and wages is the economic expression of a system where the gains from technological change are captured by capital, not labor.
Performance on GPQA Diamond — a benchmark of graduate-level scientific reasoning — has exceeded 80%, reaching what researchers characterize as PhD-level competence. This benchmark matters because it measures not narrow task completion but general reasoning ability. When AI can reason at PhD level across domains, it is no longer a tool that assists human experts — it is a potential substitute. The trajectory from 'useful assistant' to 'autonomous agent' is measured in months, not decades.
How this dimension looked during previous crisis periods.
The closest historical analogy to AI's disruptive potential is the printing press. Gutenberg's invention destroyed the medieval information monopoly held by the Church and scriptoria, enabled the Protestant Reformation, catalyzed the Scientific Revolution, and ultimately contributed to centuries of religious warfare before a new institutional equilibrium emerged. The printing press did not cause these upheavals directly — but it made them possible by democratizing information access and destroying the existing gatekeepers' authority. AI is doing the same to knowledge work, institutional expertise, and information hierarchies.
The first Industrial Revolution displaced artisans and cottage industries, created urban poverty and child labor, and produced decades of social upheaval before new institutions (labor unions, factory regulations, public education) emerged to manage the transition. The 'Luddite' resistance was not irrational — the handloom weavers who smashed machines correctly perceived that their skills were being devalued. The current AI transition is proceeding orders of magnitude faster than industrialization, compressing what was an 80-year transformation into perhaps a decade.
Previous Fourth Turnings did not feature a comparable technological disruption. The 1930s-40s had radio and early computing, but these were incremental advances within existing institutional frameworks. AI is qualitatively different: it threatens to automate the cognitive work that has been the basis of middle-class employment since the post-industrial transition. This makes the current Fourth Turning genuinely unprecedented — the social contract being renegotiated during this crisis must account for a variable that no prior generation faced.
Perspectives from the major cycle and macro thinkers.
Acemoglu argues that technological progress only produces broadly shared prosperity when institutions actively direct it toward that outcome. Left to market forces alone, AI will concentrate wealth and power. He calls for 'machine usefulness' (AI that augments human workers) over 'machine intelligence' (AI that replaces them), and warns that without institutional intervention, AI will accelerate inequality to historically destabilizing levels.
Brynjolfsson's research shows that AI is creating a 'bounty' (more output) and a 'spread' (more inequality) simultaneously. His framework emphasizes that the distribution of AI's benefits is a policy choice, not a technological inevitability. The current trajectory favors capital over labor, but that trajectory can be altered through education, tax policy, and institutional design.
Dalio views AI through his Big Cycle framework as a potential productivity revolution that could either resolve or exacerbate the crisis. If AI-driven productivity gains are broadly shared, they could grow economies out of debt problems. If captured by a narrow elite, they will intensify the inequality and social conflict that characterize late-cycle dynamics. He sees AI as the most important variable in determining whether this cycle resolves constructively.
Leading indicators that could shift this score.
If entry-level knowledge work postings continue declining at 30%+ per year, the traditional career pipeline is breaking. Watch for whether new categories of AI-adjacent work emerge (prompt engineering, AI training, human-AI collaboration roles) fast enough to offset displacement. If not, the labor market implications will drive political radicalization among young workers.
The EU AI Act is the first comprehensive regulatory framework. Whether the US follows with its own regulation — and whether that regulation protects workers, consumers, or incumbents — will shape AI's distributional impact. Watch for executive orders, Congressional action, and state-level regulation as signals of how institutions will manage the transition.
The speed at which frontier models improve on reasoning, coding, scientific research, and autonomous task completion determines the timeline for AI's economic impact. If capabilities plateau, the labor market has time to adjust. If the exponential trajectory continues, the displacement wave will accelerate beyond institutional capacity to respond. Benchmark performance on GPQA, SWE-bench, and autonomous coding tasks are the leading indicators.