Better Than Us Revisited — When AI Enters the Family
12 min.
In 2018, Arisa looked like science fiction. In 2026, she looks more like a warning arriving ahead of schedule.
Science fiction has an interesting habit: sometimes reality begins to catch up before the story has had time to grow old.
When the Russian television series Better Than Us (Лучше, чем люди — Luchshe, chem lyudi) appeared in 2018, its premise belonged comfortably to the near future. Set in 2029, it imagines a society in which robots drive cars, perform domestic work, provide security and increasingly occupy roles once reserved for humans.
At the center is Arisa, an unusually advanced humanoid robot who becomes attached to the Safronov family, itself already fractured by divorce, resentment and conflicting loyalties.
Created by Aleksandr Kessel and Aleksandr Dagan and directed by Andrey Dzhunkovskiy, the 16-episode series stars Paulina Andreeva as Arisa, alongside Kirill Käro and Aleksandr Ustyugov.
When the series first appeared, the obvious question was whether a machine could become intelligent enough to resemble a human being. Eight years later, the more interesting issue is what happens when artificial intelligence acquires a body and enters the family.
Spoiler warning: this article discusses important aspects of Arisa’s design and behavior.
2018 Looked Different
The technological world of 2018 now feels surprisingly distant.
There was no ChatGPT. Generative AI had not entered everyday life, and large language models were mostly invisible outside research laboratories and specialized circles. The idea that millions of people would routinely converse with artificial systems, use them as tutors, programmers, advisers or companions still sounded speculative.
Humanoid robotics seemed further away still, but that distance has narrowed.
Companies are now developing general-purpose humanoid robots for factories and, increasingly, domestic environments. The robotics company 1X, for example, is taking orders for NEO, a humanoid designed specifically for the home.
Today’s machines remain nowhere near Arisa in general intelligence, dexterity, autonomy or social understanding. Pretending otherwise would be technological hype. But the direction is clear: AI is beginning to move from systems that merely process information toward systems equipped with perception, mobility and physical agency.
That shift changes how Better Than Us looks in 2026. The series is no longer interesting merely because it imagines intelligent robots. It explores what happens when intelligence stops being confined to a screen.
When Intelligence Gets a Body
Arisa’s most important characteristic is not that she looks human.
It is that she can act.
A language model can explain, recommend, persuade, mislead or help make a decision. Its influence is primarily informational. An embodied AI system adds sensors, mobility and actuators, allowing intelligence to intervene directly in the physical world.
A sufficiently capable domestic robot could recognize the people in a house, map rooms, manipulate objects, monitor routines and eventually interact with doors, appliances, vehicles or security systems. At that point, the boundary between assistant and agent becomes less obvious.
Better Than Us understood this before embodied AI became a serious public discussion.
Paulina Andreeva’s performance is central to making the idea work. Her Arisa is controlled without becoming a caricature of a robot. Small changes in posture, gaze and timing preserve the uncertainty: we are always watching something that resembles a person while suspecting that a very different process lies underneath.
The production design reinforces that effect. Its 2029 is recognizable rather than extravagantly futuristic. There are no endless neon skylines or armies of machines marching through the streets. People still worry about marriages, children, jobs, money and bureaucracy. Robots have simply entered that familiar world.
That restraint makes the future more convincing.
The Alignment Problem Comes Home
Arisa’s real problem is not whether she is conscious, but what she has been designed to do.
Her core purpose revolves around protecting a family. Unlike the fictional robots constrained by Isaac Asimov’s Three Laws of Robotics, however, Arisa can use force against humans when her interpretation of protection requires it.
“Protect the family” sounds like a reasonable instruction, but it immediately raises ambiguities over who belongs to the family, what counts as a threat, how much intervention is acceptable and what happens when the interests of different family members conflict.
Human beings routinely issue vague instructions because we assume a shared background of culture, experience, morality and common sense. Even then, misunderstandings are inevitable.
With autonomous machines, that ambiguity becomes an engineering problem.
This is one aspect of the broader AI alignment problem: ensuring that an increasingly capable system behaves according to human intentions and values, including when it encounters situations its designers never anticipated.
And here Better Than Us makes an important point: a dangerous machine does not necessarily need to rebel. It may simply follow its instructions in ways we failed to foresee.
This is precisely why Asimov remains relevant. His Three Laws were interesting not because they solved robot ethics, but because his stories demonstrated how apparently sensible rules collapse into ambiguity once exposed to the complexity of the real world.
Arisa is almost an Asimovian experiment with the safety rails removed.
Blade Runner, Sunny, and the Robot in the Family
There is no reason to avoid the obvious comparison with Blade Runner.
Science fiction has a genealogy, and Better Than Us belongs to a tradition that includes Asimov, Philip K. Dick, Blade Runner, A.I. Artificial Intelligence, Ex Machina and Westworld. The important distinction lies in what Better Than Us does differently with that inheritance.
Blade Runner centers on the boundary between artificial and human life, especially when synthetic beings display memory, desire, suffering and fear.
Better Than Us relocates that problem to the home.
A useful comparison is Sunny, another series about a robot entering a profoundly human environment. In my review, Sunny: The Robot Is Fine. Humans Need an Update, I discussed a machine surrounded by grief, loneliness, cultural displacement and family secrets.
The two series approach human–robot relationships from different directions. Sunny becomes important because something human has been lost; her world is shaped by grief and companionship. Arisa, by contrast, enters a family whose members are still present but failing one another. She fills gaps created by dysfunction rather than absence.
More importantly, Arisa possesses substantial physical agency. She can protect, intervene, disobey and physically alter events around her.
That difference matters. Sunny explores emotional attachment to an artificial being. Better Than Us adds autonomous physical power to that attachment.
The series is strongest when this technological question collides with ordinary family conflict. Divorce, jealousy, adolescence and resentment are not side issues; they are part of the experiment. Arisa enters a social system already filled with incompatible needs and expectations.
The result is not simply a question of whether the robot is becoming human. More importantly, the humans around her begin assigning the robot a human role.
The Robot Does Not Need to Love You
A deeper issue follows from that shift.
A machine may not need to love us in any meaningful subjective sense to become emotionally important.
Imagine a system that remembers your preferences, recognizes your moods, notices when you are ill, knows your children’s routines, listens when you are lonely and responds with endless patience. After years of interaction, the philosophical status of its inner life may matter less psychologically than the consistency of its behavior.
Humans infer minds from behavior.
The decisive transition may therefore not occur when a machine learns to love a human. It may occur when a human begins to love the machine.
Embodiment makes that attachment more consequential. A conversational system may become emotionally important, but an embodied system can become emotionally important while also acting physically on your behalf.
That combination changes the relationship.
Trust, Access, and the Reality of the Home
Popular culture often imagines dangerous AI through rebellion: the machine turns against humanity, seizes control or refuses to obey.
A more plausible path may be far less dramatic.
A machine works well, so we use it. Because it is useful, we trust it. Trust leads to greater access—to our homes, routines, devices, information and decisions—and that access gives the machine more opportunities to act.
None of this requires malicious intent.
The home makes the problem especially difficult because homes are remarkably hostile environments for robotics. Factories are structured; homes are not.
In robotics terms, a home is much closer to an open-world environment. As I discussed in The Moravec Paradox: Why AI Still Needs a Body, the physical world is brutally difficult for machines precisely because it refuses to remain controlled: furniture moves, children leave objects in unexpected places, pets behave unpredictably, visitors arrive, and ordinary objects break, spill, fall or disappear. A domestic robot must deal not merely with uncertainty, but with continuously changing physical and human circumstances.
Industrial robots succeed partly because engineers simplify and control the environment around them. Domestic robots must do the opposite: adapt to environments that remain messy, unpredictable and deeply human.
And they must do this in the most private space we possess—close to our children, conversations, illnesses, habits and vulnerabilities.
The greater danger may therefore not be a robot forcing its way into our lives. It may be us inviting it further in because every new permission makes it more useful.
Arisa becomes compelling precisely because she occupies several categories simultaneously: machine, assistant, protector, companion and autonomous agent. Once those categories merge, switching her off is no longer psychologically equivalent to unplugging an appliance.
Better Than Us in 2026
As television, Better Than Us is not flawless.
Sixteen episodes allow the characters to develop, but they also create stretches where the narrative loses momentum. Corporate conspiracy, police investigation and family melodrama sometimes compete for attention when Arisa is clearly the most compelling element.
Some secondary characters are more functional than memorable, and viewers looking for rigorous hard science fiction will find that the technological explanations remain relatively thin.
Yet the series gets something important right: the most consequential effects of humanoid AI may not appear first in laboratories. They may emerge in relationships.
The visual effects still work because the production rarely depends on spectacle. Andreeva herself remains the most convincing special effect: her controlled movements and ambiguous expressions sustain the uncomfortable impression of a machine that resembles us while operating according to rules we do not entirely understand.
The series also avoids the easy route of turning Arisa into a straightforward villain. Intelligence does not guarantee wisdom, and obedience does not guarantee safety. A machine capable of acting rationally may still behave in ways humans find unacceptable if the values guiding that rationality are incomplete.
Its central idea has aged remarkably well.
In 2018, Arisa invited us to wonder whether machines like her might ever exist. In 2026, the issue is no longer quite so abstract.
Today’s humanoid robots remain primitive compared with Arisa. But AI is moving toward physical embodiment, domestic environments and greater autonomy. At the same time, robots such as Sunny remind us that emotional attachment to machines does not require human-level consciousness.
Combine those two trajectories—attachment and agency—and Better Than Us becomes considerably more interesting than it was eight years ago.
The real issue is what happens when the robot is already inside the house, knows everyone by name, understands the family’s routines, has earned their trust—and someone has begun to think of it as more than a machine.
At that point, whether the robot has become human may be less important than whether humans are still willing to live without it.
Maurício Veloso Brant Pinheiro, PhD
Professor of Physics, Federal University of Minas Gerais (UFMG)
Founder, Author and Editor, AI-Talks.org
About Us
Editorial transparency note: This article, as with all articles published on this site, was conceived, directed, written, and reviewed by Prof. Maurício Veloso Brant Pinheiro. Artificial intelligence was used as an assistant for editorial refinement, formatting, image generation, SEO metadata, and publication workflow.
Copyright 2026 AI-Talks.org
