A Guardian columnist argues that while fears mount that artificial intelligence could entrench a “permanent underclass,” the technology’s economics and execution look shakier than its boosters claim—and workers still have leverage. Citing predictions of near-term superintelligence and job obsolescence, the piece counters with signs of an AI bubble: investor unease, central-bank warnings on frothy valuations, and disappointing enterprise results. Studies are noted showing limited ROI from pilots and high failure rates for AI agents on basic office tasks, alongside evidence that companies that replaced staff with automation often regret it.
The article contends AI’s industrial inevitability is a myth, highlighting historical precedents where organized labor negotiated better terms amid technological change—from Japanese auto plants’ “cooperative modernization” to the WGA’s 2023 deal limiting AI’s role. With data-center costs rising and assets potentially depreciating rapidly, the author suggests the replaceable parties may be the hype merchants propelling the boom, not the workforce they claim AI will supplant.
Related articles:
Generative AI at Work: The Impact of Large Language Models on Customer Service Productivity
Artificial Intelligence and Copyright – U.S. Copyright Office
