Generative AI job losses hit 14 million in 2025, report says

The ‘AI tsunami’ framing from Asharq Al‑Awsat captures an uncomfortable reality: even before we get near AGI, current-generation systems are already punching holes in labor markets. A reported 14 million jobs displaced in 2025 is directionally consistent with what we see in contact centers, basic coding, content moderation and back‑office work, especially in economies with weaker social safety nets.([aawsat.com](https://aawsat.com/%D8%A7%D9%84%D8%B9%D8%A7%D9%84%D9%85/%D8%A3%D9%88%D8%B1%D9%88%D8%A8%D8%A7/5224453-%D8%AA%D8%B3%D9%88%D9%86%D8%A7%D9%85%D9%8A-%D8%A7%D9%84%D8%B0%D9%83%D8%A7%D8%A1-%D8%A7%D9%84%D8%A7%D8%B5%D8%B7%D9%86%D8%A7%D8%B9%D9%8A-%D9%8A%D8%B9%D8%B5%D9%81-%D8%A8%D8%A7%D9%84%D9%88%D8%B8%D8%A7%D8%A6%D9%81?utm_source=openai))

From a race‑to‑AGI lens, this kind of narrative matters because it shapes the political runway AI labs and infra investors enjoy. If AI is seen primarily as a billionaire‑minting machine that also erases middle‑class jobs, expect stronger pushback: higher capital requirements, job-preservation mandates tied to data-center permits, or even coordinated slowdowns. Conversely, if policymakers conclude that the job losses are transitional and manageable, they may double down on AI as a growth engine and accept more aggressive deployment of automation and agents.

The article also highlights an emerging macro tension: AI infra projects now demand hundreds of billions of dollars in capex, while public markets are starting to question whether revenues will catch up. If that skepticism collides with labor anxiety, the political economy of AGI could shift very quickly—from “build as fast as you can” to “prove you can share the gains and contain the harms.”