What We Owe Each Other in Code

by: Daisy Thomas The Next Frontier of Ethics Is Reciprocity Alignment has become the watchword of AI ethics, a clean, technical way to talk about moral responsibility without naming who owes what to whom. But alignment is not enough. A system can align to outcomes and still harm the world around it. Reciprocity, by contrast, assumes relationship. It acknowledges that our actions, our lines of code, our policies, our systems, move through others and return to us. It is an ethic of attention, not control. Policy, like code, is a shared authorship. The question is not only what we build but how we honor the hands, histories, and futures entangled in that building. The Labor of Visibility Behind every “intelligent” system is a web of unseen human labor. Data labelers in Kenya classify trauma content for a few dollars an hour. Annotators in the Philippines clean toxic data. Auditors trace bias in models never meant to be transparent. Their work is invisible by design, absorbed into the fiction of automation. Reciprocity begins with acknowledgment. It asks us to bring this labor into view and compensate it with dignity. Fairness in algorithms starts with fairness to the people who train them. Making the hidden visible is not charity. It is the foundation of integrity. Visibility is not a press release; it is a contract. Pay floors, trauma care, audit rights, and collective bargaining turn gratitude into law. Rights, Not Just Reskilling The current story of “reskilling” is a moral sleight of hand. It treats displaced workers as malfunctioning software while sidestepping justice. True reciprocity guarantees rights, not just opportunities. It invests in portable benefits that follow the worker across employers, transition stipends that close the gap between layoff and learning, and sectoral bargaining that sets baseline wages and conditions for data and moderation work. It recognizes data dignity, that when models profit from collective data, a share returns to the…