Part of an occasional series: ideas from outside the usual solarpunk reading list.
Almost every public conversation about robots and AI reaches, sooner or later, for the same comparison: will it replace us? Is it as smart as us? Will it deserve rights the way we do? Kate Darling — a robot ethicist at the MIT Media Lab — argues in her 2021 book The New Breed that this whole comparison is the source of the confusion, not the path out of it. Her fix: stop comparing robots to humans, and start comparing them to domesticated animals instead.
Why the human comparison backfires
Darling’s argument, stated plainly in her introduction: measuring robots against human capability invites a category error. It frames every robot as a deficient human — not as good at reasoning, not as good at empathy, not as good at whatever we happen to be measuring — rather than as something genuinely new that doesn’t need to clear a human-shaped bar to be useful, ethically significant, or worth taking seriously. This comparison is also, she argues, the direct source of the „uncanny valley” problem — the specific unease people feel around robots that look almost human but not quite — and of the recurring moral panics about robots „taking our jobs” or robots „having rights,” debates that stay confused because they’re implicitly asking whether a robot is human-equivalent rather than asking what it actually is.
Her alternative: we have millennia of accumulated practice incorporating a completely different category of non-human beings into work, transport, companionship, and even warfare, without ever needing them to be human-equivalent to matter — domesticated animals. That existing body of legal, ethical, and practical precedent, Darling argues, is a far better starting toolkit for thinking about robots than anything built from scratch around a human-comparison frame.
Three movements
The book is organized in three parts.
Work, weaponry, responsibility. Darling traces the long history of animal labor — oxen, camels, war dogs, pigeon-guided bombs, even military dolphins — to argue that robots, like working animals throughout history, should supplement human labor rather than simply replace it. The most immediately useful piece here is legal: the „goring ox” tradition, a body of liability law running from the Code of Hammurabi through English common law to today’s dog-bite statutes, gives a ready-made, non-exotic framework for assigning responsibility when an autonomous animal — or, by extension, an autonomous robot — causes harm, without needing to invent robot personhood or attempt to program a full system of machine ethics from first principles. We’ve already solved a structurally similar problem, repeatedly, across very different legal traditions; Darling’s argument is that we should use that accumulated solution rather than reinvent it.
Companionship. This is the heart of the book’s most original contribution. Darling surveys research on why people anthropomorphize machines — including psychologists Byron Reeves and Clifford Nass’s „media equation” work, showing people unconsciously apply social rules to technology that in any literal sense clearly doesn’t warrant them, and documented real-world cases like people naming their Roomba vacuum or holding funeral services for decommissioned AIBO robot dogs at Buddhist temples in Japan. From this she extracts practical design principles for social robots: aim for less, not more, humanlike realism, since human-adjacent-but-imperfect design is exactly what triggers uncanny-valley discomfort, while a design that reads clearly as animal-like or creature-like (rather than almost-human) tends to sidestep that discomfort while still supporting a genuine emotional bond. Her central claim: our demonstrated capacity to love animals — fully knowing they’re „just animals,” not people — is a direct, workable model for how we might come to relate to robots, fully knowing they’re „just machines.”
A predictive model for robot rights. The book’s final movement uses the history of the Western animal-rights movement — its gradual expansion, its persistent controversies, its incomplete and contested legal victories — as a forecast for what a coming robot-rights conversation might actually look like: messy, partial, driven as much by public sentiment and specific high-profile cases as by clean philosophical argument.
What the book is not trying to do
It’s worth being precise about the book’s actual scope, because it’s easy to read more into it than it claims. Darling’s focus is almost entirely on how humans should think about, design, and legally treat robots — not on whether robots themselves might be owed anything, in the sense of having interests or experiences that matter morally in their own right. That second question — is the AI itself a moral patient, something that can be wronged, not merely something we might be wrong to describe carelessly — is explicitly outside the book’s scope. Readers interested in that harder, adjacent question might find our companion piece on Jeff Sebo’s work on precautionary moral status a natural next read; the two arguments pair well precisely because they’re answering different halves of the same larger question.
Why the animal comparison works better than it might sound
The strongest evidence for Darling’s approach is that it’s already been tested, informally, at scale: it’s the comparison ordinary people reach for spontaneously, long before any ethicist tells them to. We already talk about robot vacuums the way we talk about pets. We already have visceral reactions to footage of robots being kicked or knocked over that read as something closer to concern for an animal than concern for a toaster. Darling’s contribution isn’t inventing this instinct — it’s naming it, giving it historical and legal grounding, and arguing we should build our institutions, our design principles, and our regulatory frameworks around the comparison people are already making, rather than around a human-equivalence framing that keeps generating confusion precisely because it doesn’t match how people actually relate to these machines.
Did we get something wrong?
This piece is drawn from our own reading notes, not a translation of the book. If something above misrepresents the argument, or just sounds off, tell us — publicly or privately, whichever you’d rather.
Want the fuller picture?
This is a survey, not the whole book — including more of the goring-ox legal history and the design-principles material. The deep-dive version is here: deep-dive PDF.
What should we read next?
This is one piece in a loose series drawn from our own research reading. If there’s a thinker, a book, or a tradition you think belongs in a series like this, we want to hear it.
