Oxford researchers say public deserves a 'happiness dividend' from AI boom
Oxford Internet Institute researchers argue in a new study that citizens who fund AI through data, money and public resources should get technology that demonstrably improves wellbeing.
Summary
The Oxford Internet Institute at the University of Oxford announced on 21 October 2026 a paper arguing that the public should receive a happiness dividend from the AI boom. The paper appears in Royal Society Open Science and concerns happiness in EU technology policy as a way to repay public subsidies for AI. Sandra Wachter leads a team that also includes Brent Mittelstadt, Daria Onitiu and Netta Weinstein.
According to the institute, the argument is that citizens are in effect shareholders in AI. They provide money, data and public resources, including funding, tax breaks, land, water, energy and infrastructure. The researchers say such AI should be judged on more than growth and risk reduction, and that supported companies should show how their technologies will improve well-being.
The proposal is to add happiness as a third aim of technology policy, on the view that reducing harm is a different thing from delivering social benefit. Using psychological research, the authors name three foundations of happiness: relatedness, meaning real connections to people and nature; competence, meaning the chance to build skills and reach goals; and autonomy, meaning the freedom to choose in line with personal values. They warn that some AI uses may erode these. Chatbots imitating human emotion could lower the value placed on human connection, dependence on generative AI could weaken skills, AI built to sway users could hinder independent choices, and data centre growth could squeeze access to land and water.
Under the approach, supported companies would forecast how their products raise happiness and be held to account for that. The institute says the principle, though developed for EU and UK policy, applies more widely.