The Uncomfortable Truth: Is Generative AI Actually Hurting the Labor Market?
Published on 13.01.2026
Generative AI and the Labor Market: Following the Data
TLDR: The AI boom has captured 92% of U.S. GDP growth while labor's share of economic output has hit a record low. After three years of generative AI, job creation is anemic, tech layoffs are surging, and the "revolution" isn't creating the jobs it was supposed to.
Let me walk through some uncomfortable numbers that don't fit the narrative we've been hearing about AI transformation.
In the final three months of 2025, the U.S. economy added negative 67,000 jobs. Labor's share of GDP dropped to 53.8% in Q3 2025—a record low going back to 1947. Meanwhile, AI investments accounted for nearly 92% of U.S. GDP growth in the first half of 2025. Consumer sentiment sits near generational lows.
Something doesn't add up here. If this were truly a technological revolution comparable to electrification or the internet, wouldn't we expect to see job creation, not destruction? The 2025 numbers tell a stark story: highest Q4 layoffs since 2008, lowest year-to-date hiring since 2010. Tech sector alone saw 170,630 layoffs—up from 95,000 in 2024.
The problem isn't that AI doesn't work. It clearly does, especially in coding where productivity gains are measurable. The problem is the distribution of those gains. BigTech companies are thriving—their capex on AI infrastructure is exploding, their earnings are solid, their stock prices are elevated. But that wealth isn't flowing to workers.
What we're seeing looks more like "Tycoon Capitalism" than broad-based technological progress. The AI infrastructure buildout benefits the companies building it and the wealthy exposed to their equities. Meanwhile, job redesigns favoring AI workflows mean fewer positions, entry-level opportunities are shrinking, and middle management roles are being purged.
Here's what the optimists are missing or avoiding: there's no clear mechanism for how AI job creation is supposed to happen. The usual response is "it takes time for new job categories to emerge"—but after three years, we should at least see early indicators. Instead, healthcare jobs are masking what would otherwise be net negative job growth.
The immigration dimension makes this worse. Nearly 80% of billion-dollar companies had an immigrant founder or immigrant in key leadership. Current policy has slowed immigration by 80-90% compared to 2023-2024 peaks. Combined with an aging population and low fertility rates, the U.S. doesn't have the internal talent pool to fill highly technical AI roles even if they were being created.
For architects and technology leaders, there's a strategic question here: are you building AI systems that augment your teams, or are you participating in workforce elimination dressed up as "efficiency"? The difference matters—not just ethically, but practically. Teams reduced too aggressively lose institutional knowledge and adaptive capacity.
The data suggests that generative AI's overall ROI for society—not just for tech companies and their investors—may be net negative three years in. The impact on culture, trust in institutions, rule of law, and human leadership has been decidedly negative. The productivity gains in coding haven't scaled to other knowledge work the way enthusiasts predicted.
This isn't an argument against AI. It's an argument for honesty about what we're seeing. If AI is creating value, we should be able to measure it in the labor market, in real wage growth, in broadly distributed prosperity. Instead, we're seeing wealth concentration, job loss, and a K-shaped economy accelerating.
The people making money from AI are telling us AI is revolutionary. The labor market data tells a different story.
Key takeaways:
- Labor's share of GDP at record low (53.8%) while AI captures 92% of GDP growth
- Tech layoffs surged to 170,630 in 2025, up from 95,000 in 2024
- Net negative job creation in Q4 2025 (-67,000 jobs)
- Healthcare jobs are masking otherwise negative overall job growth
- Immigration slowdown compounds talent shortage in technical roles
Tradeoffs:
- AI efficiency gains for companies come at the cost of workforce stability
- Rapid AI infrastructure investment delivers stock gains but suppresses labor share of GDP
- Optimizing for AI-native workflows reduces hiring while increasing technical capability