The economic alignment problem of artificial intelligence
For AI researchers and policymakers, this paper reframes AI alignment as an economic issue, but the proposals are conceptual and lack empirical validation.
The paper argues that the AI alignment problem is also an economic alignment problem, as growth-oriented economic systems increase AI risks. It proposes post-growth concepts like satisficing, Doughnut economics, and resource caps to reduce risks, and suggests governance reforms treating AI as a commons.
Artificial intelligence (AI) is advancing exponentially and is likely to have profound impacts on human wellbeing, social equity, and environmental sustainability. Here we argue that the "alignment problem" in AI research is also an economic alignment problem, as developing advanced AI within a growth-oriented economic system is likely to increase social, environmental, and existential risks. We show that post-growth research offers concepts and policies that could address the economic alignment problem and substantially reduce AI risks, such as by replacing optimisation with satisficing, using the Doughnut of social and planetary boundaries to guide development, and curbing systemic rebound with resource caps. We propose governance and business reforms that treat AI as a commons and prioritise tool-like autonomy-enhancing systems over agentic AI. Finally, we argue that the development of artificial general intelligence (AGI) requires new economic theories and models, for which post-growth scholarship provides a strong foundation.