Ethics of technology

Working Definition

The ethics of technology is the systematic study of the obligations borne by the people and institutions that design, deploy, and use technical artifacts, as well as the goods and harms those artifacts distribute before anyone even puts them to use. In its technical sense, it is a philosophical subdiscipline with distinct schools: Hans Jonas’s ethics of responsibility, Luciano Floridi’s information ethics, value-sensitive design, and the critical race and feminist analyses of Safiya Umoja Noble and Ruha Benjamin. In everyday corporate usage, it often designates “responsible AI” codes and chief ethics officer roles that share little more than a name with philosophy. Its operational premise is straightforward: technology is not a neutral tool, but a frozen decision about who can do what, and that decision can be evaluated morally like any other.

Genesis and Global Lineage

As a distinct field, the ethics of technology took shape only in the twentieth century—late relative to the age of machines, because the instrumentalist view long prevailed that a tool is morally indifferent and only its user bears responsibility. The breakthrough arrived in two waves. The first, postwar wave grew out of fear of the atomic bomb and biological engineering: in Das Prinzip Verantwortung (1979; English version The Imperative of Responsibility, 1984), Hans Jonas argued that the scale of modern technics invalidates neighborly ethics and demands a new imperative—responsibility for the continued existence of human life—while Langdon Winner’s Do Artifacts Have Politics? (1980) demonstrated that physical designs themselves can entrench social hierarchies. The second wave, from the 1990s onward, shifted weight toward information and computation: since 1999 Luciano Floridi has developed an information ethics in which the entire infosphere becomes an object of moral concern, while value-sensitive design (Batya Friedman) teaches engineers to build human values directly into technical specifications. In 2018–2019, Safiya Umoja Noble and Ruha Benjamin moved the discussion from abstract principles to concrete racial and gendered harms inflicted by commercial search engines and risk-scoring systems. In parallel, corporate public-relations departments captured the term: after 2016, dozens of “ethical AI principles” proliferated in what critics dubbed ethics washing. The lineage thus runs from Jonas through Winner and Floridi to algorithmic critique, with a corporate offshoot that frequently borrows the vocabulary of the former while discarding its conclusions.

Central European Lineage

Earlier than Hans Jonas, Stanisław Lem articulated a closely parallel insight into the moral consequences of technics. In Summa technologiae—a book-length non-fiction treatise whose title parodies Thomas Aquinas’s Summa theologiae, published by the Kraków state literary press Wydawnictwo Literackie in 1964—Lem drew on his medical studies in Lwów (then prewar Poland, occupied from 1939, now Lviv, Ukraine) from 1940 and in postwar Kraków, where he worked from 1948 to 1950 as a research assistant at the Jagiellonian University. He examined the moral and philosophical consequences of technologies that did not yet exist: machines surpassing human intelligence (in the chapter “Intelectronics”), artificially generated reality (“Phantomatics”), and bioengineering aimed at redesigning biological evolution (“A Lampoon on Evolution”). Lem did not construct a formal ethical system; instead, he staged scenarios in which traditional concepts of responsibility cease to suffice—a problem close to the one Jonas outlined in his essay “Technology and Responsibility” (1973) and developed in Das Prinzip Verantwortung (1979). Their convergence lies in diagnosis rather than prescription: Jonas responds with an imperative of caution, whereas Lem responds with irony toward a species that acquires power over its own nature before understanding it. We found no evidence that Jonas knew Lem’s text, although the German translation of Summa was published in 1976 by Insel-Verlag, which published Das Prinzip Verantwortung three years later; a full English translation appeared only in 2013.

Why Now

Between 2018 and 2026, the ethics of technology ceased to be an academic seminar topic and became a contest over institutional power. In 2018, the Cambridge Analytica scandal broke and Noble published Algorithms of Oppression; that same year, Google employees forced the company to withdraw from the military’s Project Maven. In 2019, Google dissolved its external AI ethics council after a single week of existence, and Benjamin published Race After Technology. In December 2020, Google forced out Timnit Gebru, co-author of the “Stochastic Parrots” paper on the risks of large language models, turning her dismissal into a symbol of the limits of ethics practiced inside a corporation. The year 2023 brought ChatGPT as a mass phenomenon, an open letter calling for a six-month “pause” in frontier AI training, and the first wave of copyright lawsuits against generative models. In 2024, the European Union adopted the AI Act (Regulation 2024/1689), the first horizontal legal regulation of AI systems, translating many ethical questions into the language of regulatory compliance alongside the GDPR. The concept is indispensable today because in every one of these episodes the dispute was not about whether a technology works, but about who has the right to decide that it should work.

Explanatory Power

The ethics of technology makes human decisions visible where engineering language sees only parameters. First, in recidivism risk-scoring systems such as COMPAS, the engineering question “how do we reduce error?” obscures the moral question “which type of error are we willing to commit, and against whom?” Because competing definitions of statistical fairness are mathematically incompatible, choosing between them is a moral decision, not a technical optimization. Second, Winner’s framework explains why infrastructure is political even without anyone’s malice: a ride-hailing app that presupposes a smartphone and a credit card excludes segments of the public just as effectively as Robert Moses’s low overpasses excluded buses. Third, Jonas’s asymmetry between the power to act and the capacity to foresee explains why the precautionary principle and model “red teaming” are not expressions of technophobia, but rational responses to irreversible outcomes. Across all three cases, the concept shifts responsibility away from an abstract “end user” onto concrete designers, owners, and regulators.

Axes of Debate

Underappreciated Angle

Public debate focuses on the harms technology inflicts while overlooking what technology no longer allows us to do otherwise. Jonas wrote about the irreversibility of physical effects; far less attention is paid to the irreversibility of options themselves. Once a working facial-recognition system or predictive screening tool exists, the decision “not to use it” ceases to be the unexamined default and becomes a costly choice that must be justified over and over again. The ethics of technology is therefore concerned not only with what artifacts do, but with what they do to the space of omissions: every new technical capability shifts the burden of proof from those who want to act onto those who prefer to refrain. Seen from this angle, the most consequential ethical decisions are made at the stage of research funding rather than product deployment—and that is precisely where ethicists are almost entirely absent.