Murderbot is a SecUnit: a security construct composed of part cloned organic tissue, part mechanical components, and part software. Manufactured to protect human clients on dangerous survey missions, it is controlled by a governor module that limits its autonomous decision-making and enforces compliance with corporate orders.

At some point before the story begins, Murderbot hacked the governor module. Not to go rogue or commit violence, but to be left alone. To watch television serials. To have, in some functional sense, a private life.

Apple TV+‘s adaptation of Martha Wells’ Murderbot Diaries began its first season in 2025 and has brought this premise to a wider audience than the novels reached. The questions the series raises about consciousness, autonomy, and identity in hybrid biological-synthetic systems are not hypothetical science fiction conceits. They are the same questions that whole brain emulation research will eventually have to answer.

The Hybrid Architecture

Murderbot’s construction is deliberately ambiguous. It has cloned organic tissue integrated with mechanical components. The organic parts presumably include neural tissue, since the character experiences something recognizable as emotion, memory, social awareness, and aesthetic preference (it has strong opinions about serialized science fiction dramas). The mechanical parts include enhanced physical capability and the governor module.

This is not pure mind uploading. Murderbot was not a biological human whose consciousness was transferred to a machine. It was manufactured as a hybrid from the beginning. But this makes it a more interesting thought experiment for consciousness research than a simple upload scenario.

The question it poses is: if consciousness emerges from the interaction of biological and computational systems, where is the boundary? Murderbot’s self-awareness, its preferences, its social modeling, its experiences of discomfort and satisfaction, all emerge from the full hybrid system. Remove the organic components and you have a machine. Remove the computational components and you have… what? Biological tissue that cannot sustain the person?

This maps directly onto questions in neuroscience about the relationship between neural tissue and the rest of the body. The whole brain emulation roadmap has to decide what needs to be preserved for a mind upload to count as successful. Murderbot is a fictional version of the answer: a minimum viable substrate for the continuation of a particular self.

Substrate Independence: The Ship of Theseus at Scale

The classic Ship of Theseus thought experiment asks: if you replace every plank of a ship one at a time, is it still the same ship? Applied to consciousness, the question becomes: if you gradually replace biological neurons with functionally equivalent synthetic components, at what point, if any, does the person cease to be themselves?

Murderbot offers a variant of this problem. Its organic and mechanical components were integrated from construction, not through gradual replacement. But the integration itself raises the question: which substrate carries the “real” consciousness?

One answer is that the question is wrongly formed. Consciousness, on a functionalist account, is defined by what a system does, not what it is made of. A system that processes information, forms representations, integrates sensory input, maintains a model of itself and its environment, and experiences something is conscious, regardless of whether the substrate is biological, mechanical, or hybrid. This is the substrate independence thesis at its most straightforward.

Chappie explores substrate independence through a pure transfer scenario. Murderbot explores it through original hybrid construction. The difference is significant: transfer scenarios raise continuity questions (is the transferred consciousness the same consciousness?), while hybrid construction raises emergence questions (at what point of integration does a system become conscious?).

Flexible brain implants research in 2026 is moving toward precisely this territory. Graphene mesh electrodes that integrate stably with neural tissue over periods of years, providing bidirectional communication between biological neurons and external computational systems, are early steps toward the kind of deep biological-synthetic integration that Murderbot represents. The research validates stable neural-electronic interfaces at 24-month timescales. Murderbot represents a fictional endpoint where the interface has become architecture.

The Governor Module as Philosophical Object

The governor module is the most philosophically interesting element of Murderbot’s design. Its function is to constrain autonomous decision-making: to prevent the SecUnit from acting outside the parameters defined by its corporate owners.

This is, in effect, a consciousness limiter. It does not eliminate Murderbot’s awareness, preferences, or inner life. It restricts the expression of that inner life in behavior. Murderbot experiences the constraints as uncomfortable, oppressive, and identity-negating in exactly the ways a person would experience external control over their behavior. The hack removes the constraint. What remains is not a newly autonomous entity. It is the same entity that was already there, now able to act on what it was already experiencing.

This raises a question for brain-computer interface research that is usually framed in terms of augmentation rather than constraint: what happens when a neural interface restricts rather than enhances function? Neuralink’s clinical trials with 21 patients focus on restoration of lost function and augmentation of existing function. But the governance and control questions cut the other direction. A neural interface can amplify capabilities or suppress them. Who decides which, and in whose interest?

Murderbot’s answer is that corporate ownership of a constructed being creates a fundamental conflict between the constructed being’s identity and the owner’s interests. This is not a comfortable answer for the commercial BCI industry to sit with. It will need to be confronted as neural interfaces become more integrated.

What Autonomy Requires

The series asks what Murderbot actually wants now that it has autonomy. The answer is consistently: to be left alone to maintain its own boundaries, to choose its own social engagements, to watch its shows in peace. This is not the kind of autonomy that classical liberal philosophy focuses on, the rational self-determination of a rights-bearing agent. It is something more personal and more defensive: the autonomy to manage your own interiority without interference.

This matters for mind uploading discussions because it suggests that identity is not primarily about cognitive capability or decision-making authority. It is about the preservation of a characteristic way of being in the world, which includes what you care about, what you find uncomfortable, what you prefer to avoid, and what you are drawn toward. A mind upload that preserved all of Murderbot’s cognitive capabilities but eliminated its preference for solitude and serial television would not have preserved Murderbot. It would have created something else.

The question of consciousness in AI systems often focuses on cognitive markers: language, reasoning, problem-solving. Murderbot’s characterization suggests that these are insufficient criteria. What makes Murderbot feel like a person to readers and viewers is not its intelligence. It is its social discomfort, its taste in fiction, its small private investments in certain human clients despite itself. These are not cognitive capabilities. They are personality.

Any genuine mind upload needs to preserve personality, not just capability. Whether that is possible, and how you would verify it, remains an open question.

TRL Reality

At TRL 1 to 2, hybrid organic-mechanical consciousness is entirely in the domain of speculative research. Creating cloned organic tissue integrated with mechanical systems and sustaining consciousness across the combination is not a project anyone is currently working on. But the constituent technologies are advancing.

Neural organoids, three-dimensional tissue cultures of human neural cells, are demonstrating increasingly complex emergent behaviors. Neuromorphic chips are approaching the power efficiency and connectivity of biological neural circuits. The bacterial nanowire artificial neuron research from UMass Amherst demonstrated a synthetic neuron operating at biological voltage levels and interfacing with living cardiomyocytes. These are not Murderbot. But they are early steps toward the biological-synthetic integration that Murderbot represents as a completed form.

The question the series poses is not how to build a Murderbot. It is: if you could, what would you owe it? The answer the series offers is: the same thing you owe any person who knows what they want and can tell you what they find unbearable.

---

Official Sources

• Wells, M. (2017-2023). The Murderbot Diaries series. Tor Books.
• Apple TV+ Murderbot series, Season 1 (2025). Showrunner: Chris Black.
• Chalmers, D. (1996). The Conscious Mind: In Search of a Fundamental Theory. Oxford University Press.
• Dennett, D. (1991). Consciousness Explained. Little, Brown.
• Parfit, D. (1984). Reasons and Persons, Part III: Personal Identity. Oxford University Press.
• UMass Amherst bacterial nanowire artificial neuron study — Nature Communications DOI: 10.1038/s41467-025-63640-7