AI Safety Should Prioritize the Future of Work

AI safety discussions are too narrow: Current AI safety efforts focus on content filtering, manipulation prevention, and cybersecurity risks, but largely ignore the significant economic and social impacts on workers and the future of employment—a gap the authors argue must be addressed urgently.

AI threatens long-term worker prosperity and equality: The paper highlights that AI's structural changes to labor markets will have serious "intertemporal impacts" on human livelihoods, with particular concern that these changes will exacerbate income inequality across society over time.

Current AI development practices resemble exploitative monopolies: The authors contend that the closed-source approach used by major AI companies functions as "rent-seeking behavior," exploiting resources while stifling creativity in labor and concentrating innovation power in few hands rather than distributing benefits broadly.

Fair compensation for data use requires new legal frameworks: The paper recommends implementing "collective licensing" and robust international copyright protections to ensure workers and creators are fairly compensated when their data is used to train AI models—addressing a currently unregulated practice.

Global AI governance must prioritize workers: Rather than treating AI safety as purely a technical problem, the authors advocate for a "pro-worker framework of global AI governance" designed to promote shared prosperity, economic justice, and meaningful work with human agency.